chore(release): auto-bump version to 0.66.68 — earned by # (minor: converged release (push-to-master bump))

Promotion record: (unknown)

.gitea/workflows/ci.yml

@@ -322,3 +322,25 @@ jobs:
           echo "::error::Required CI checks not green — refusing to publish required-checks success"
           echo "lint=${{ needs.lint.result }} unit-tests=${{ needs['unit-tests'].result }} canary=${{ needs.canary.result }} best-practices=${{ needs['best-practices'].result }} documentation=${{ needs.documentation.result }} type-check=${{ needs['type-check'].result }} gitea-offline=${{ needs['gitea-offline'].result }}"
           exit 1
+
+  notify-event-bridge:
+    # T16 §16C: post the CI completion WAKE to the local event bridge so
+    # event-driven waiters (deploy.yml → ci_event_wait.py) resume on the
+    # event instead of sleep-polling. ALWAYS runs (success or failure —
+    # a red CI must wake the waiter just as loudly as a green one). The
+    # wake is best-effort and never decides the verdict: the waiter
+    # re-reads the canonical Gitea state via deploy_qualify.py.
+    name: CI Pipeline / notify-event-bridge
+    runs-on: rune-ci-desktop
+    needs: [required-checks]
+    if: always()
+    steps:
+      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
+        with:
+          fetch-depth: 1
+      - name: Wake event bridge (wake only — never completion truth)
+        env:
+          CI_SHA: ${{ gitea.sha }}
+          CI_RESULT: ${{ needs.required-checks.result }}
+          RUNE_EVENT_BRIDGE_URL: ${{ vars.RUNE_EVENT_BRIDGE_URL || 'http://127.0.0.1:8791' }}
+        run: python3 scripts/notify_event_bridge.py

.gitea/workflows/deploy.yml

@@ -19,28 +19,38 @@ jobs:
         with:
           fetch-depth: 0
 
-      - name: Wait for CI qualification on this exact SHA
+      - name: Wait for CI qualification on this exact SHA (event-driven, T16 §16C)
+        # T16 §16C/§16D: the old sleep-30 → query → repeat loop is gone.
+        # The waiter registers a correlation on the W/R event bridge and
+        # yields in select() until ci.yml's notify-event-bridge job posts
+        # the completion WAKE; the verdict is then re-read from canonical
+        # Gitea (deploy_qualify.py). The 900s deadline is a FAILURE
+        # DETECTOR (lost-event recovery), never the scheduler.
         env:
           GITEA_SERVER: ${{ gitea.server_url }}
           GITEA_REPO: ${{ gitea.repository }}
           GITEA_TOKEN: ${{ secrets.ROBOT_TOKEN }}
+          # Shared bridge data root: the deploy step runs as the runner
+          # user while the daemon runs as nick — the rendezvous state
+          # (waits/events/resumes/leases) must be shared. §16C live
+          # finding: defaulting to the invoking user's home made the
+          # controller unable to create its FIFO (PermissionError) and
+          # the deploy failed fast+truthfully on both 0.66.63 pushes.
+          RUNE_WR_ROOT: /mnt/workdrive/wr-bridge
+          RUNE_WR_CODE: /home/nick/.local/share/mimo-wait-resume
         run: |
           DEPLOY_SHA=$(git rev-parse HEAD)
           export DEPLOY_SHA
-          echo "Qualifying $DEPLOY_SHA"
-          DEADLINE=$((SECONDS + 900))
-          while [ $SECONDS -lt $DEADLINE ]; do
-            STATUS=$(python3 scripts/deploy_qualify.py)
-            echo "required-checks: $STATUS"
-            if [ "$STATUS" = "success" ]; then exit 0; fi
-            if [ "$STATUS" = "failure" ]; then
-              echo "::error::CI failed on $DEPLOY_SHA — deployment refused"
-              exit 1
-            fi
-            sleep 30
-          done
-          echo "::error::CI qualification timeout on $DEPLOY_SHA — deployment refused"
-          exit 1
+          echo "Qualifying $DEPLOY_SHA (event-driven)"
+          # READINESS BEFORE ACTION (§16U): fail with a clear message if
+          # the bridge rendezvous is not writable in THIS user's context,
+          # instead of a stack trace at FIFO creation time.
+          if ! touch /mnt/workdrive/wr-bridge/resumes/.write-probe 2>/dev/null; then
+            echo "::error::RUNE_WR_ROOT not writable by $(id -un) — bridge data root misconfigured"
+            exit 1
+          fi
+          rm -f /mnt/workdrive/wr-bridge/resumes/.write-probe
+          python3 scripts/ci_event_wait.py --sha "$DEPLOY_SHA" --deadline 900
 
       - name: Stage, verify, promote (monotonic)
         env:

.gitea/workflows/maintenance.yml

@@ -22,9 +22,12 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - name: Run all canaries
+        # T16 §16Q reconciliation: the glob looped shared/canary_*.py —
+        # real canaries live in src/rune/ (canary_native.MODULES is the
+        # registry authority). The old job silently no-opped.
         run: |
           FAILED=0
-          for canary in shared/canary_*.py; do
+          for canary in src/rune/canary_*.py; do
             if [ -f "$canary" ]; then
               echo "Running $canary..."
               python3 "$canary" || FAILED=$((FAILED + 1))

README.md

@@ -1,6 +1,6 @@
 # Rune
 
-**Current release: v0.66.58**
+**Current release: v0.66.68**
 
 **Rune is a persistent personal agent designed to accumulate experience, develop reusable competence, and remain recognisably the same agent even when its underlying foundation model changes.**
 

config/umbrellas.json

@@ -1,7 +1,7 @@
 {
  "_comment": "GENERATED by rune.spine \u2014 compatibility view for legacy umbrella consumers. DO NOT EDIT: structural truth lives in store/spine/current.json; edits here are flagged by spine drift.",
  "_generated_by_spine": true,
- "generated_at": "2026-09-14T06:08:20Z",
+ "generated_at": "2026-09-15T16:31:19Z",
  "umbrellas": {
   "capability": {
    "description": "Tool selection and execution authority",
@@ -17,6 +17,7 @@
     "local_tool_runtime",
     "message_queue",
     "openai_compat_transport",
+    "provider_context",
     "recall_uplift",
     "route_risk_flags",
     "tool_executor",
@@ -219,19 +220,26 @@
     "cognitive_budgeter",
     "complexity_scorer",
     "construction_activation",
+    "construction_commands",
     "construction_decision",
     "construction_depth",
     "construction_interface",
     "construction_mode",
     "construction_pattern",
     "construction_patterns_builtin",
+    "construction_projection",
     "context_broker",
     "context_pressure",
+    "drift_resume",
+    "event_wait",
+    "failure_recovery",
     "layout_contract",
+    "programme_continuation",
     "release_notes",
     "salience_gate",
     "scaffold",
     "spine",
+    "stage_guards",
     "structural_store",
     "toolchain",
     "tui_backlog",

docs/STRUCTURAL_MAP.md

@@ -0,0 +1,71 @@
+# Rune Structural Map
+
+_Maintained structural projection — the fast re-entry surface (T16 §16N).
+Companion to the Spine store (`store/spine/current.json`, structural
+authority) and `store/evidence/t16/16M_inventory.md` (full classification
+record). Update this map when authority or canonical routes change; do
+not copy the repo here._
+
+## What is Rune?
+
+A self-hosting coding agent: TUI client over an operator core, with a
+structural self-model (Spine), a construction authority (Scaffolding),
+an event-driven wait/resume bridge (W/R), a research pipeline, and a
+fail-closed release/deployment pipeline that builds, tests, releases,
+and deploys Rune itself to `/mnt/workdrive/rune-code/current`.
+
+## Where things happen (canonical routes only)
+
+| Concern | Authority | State |
+|---|---|---|
+| Structural self-model | `src/rune/spine.py` | `store/spine/current.json` |
+| Construction (build lifecycle) | `src/rune/scaffold.py` + `construction_*.py` | scaffold store (`<dir>/.rune/scaffold`) |
+| Builder sessions | `scripts/t15_builder_session.py` (event-driven successor; T14 is legacy) | contract + wait records |
+| Provider/model selection | `rune_operator` final decider; ambient context via `provider_context.require_model` | `state/model_routing_profiles.json` |
+| Wait/resume (any long op) | `src/rune/event_wait.py` + W/R daemon (out-of-repo snapshot at `~/.local/share/mimo-wait-resume`, data root `/mnt/workdrive/wr-bridge`) | waits/events/leases/resumes JSON |
+| CI qualification verdict | `scripts/deploy_qualify.py` (two legs, fail-closed) | Gitea commit status + runs |
+| Release/deploy | `scripts/deploy_stage.py` → `deploy_promotion.py` (monotonic, flock) | deploy root + `current`/`last-good` symlinks |
+| Research | `scripts/research_task_tree.py` (authority; labels are projection) | `research/issue-<N>/` |
+| Task completion truth | `verification_layer.py` + `orchestration/conversation_state.py` | durable event store + checkpoints |
+| Routing (task class) | `src/rune/adaptive_router.py` | `routes/route.jsonl` |
+| Canary registry | `src/rune/canary_native.MODULES` | observed by Spine |
+| Structural health | `henxels_checks.py` + `henxels.yaml` | repo root |
+
+## Known non-canonical / duplicate routes (do not extend; reconcile via 16Q)
+
+- `build_mode.BuildManager` (legacy /build) — superseded by builder_bridge
+  + event-driven sessions; still wired.
+- `scripts/t14_builder_session.py` — superseded by t15 event-driven driver.
+- Provider selection entry points ×5 (`model_routing.rank`,
+  `model_connect route`, `model_registry get`, `provider_registry`,
+  `work_manager`) — operator is the de-facto final decider.
+- Task-ish state ×4 (task-tool DB, `task_persist` persisted.json,
+  `memory_store.working`, orchestration event store).
+- Maintenance ledgers ×3 (scaffold maintenance_item, tui_backlog, drift_ledger).
+
+## Unfinished / blocked / deprecated
+
+- programme_continuation.py: implemented + tested but UNWIRED (no
+  production caller) — the "continue problem" machinery awaits routing.
+- maintenance.yml canary-suite job: loops `shared/canary_*.py` — wrong
+  dir (real canaries in `src/rune/`), job likely no-ops (LEGACY).
+- daemon out-of-repo: the W/R daemon is deployed from
+  `~/.local/share/mimo-wait-resume` with its own git snapshot —
+  packaging debt, no CI coverage of the deployed copy.
+- Substrate: NOT BUILT. Scaffolding exposes the seam
+  (`construction_projection.py`, `construction_commands.py`); mock
+  client proves compatibility (`tests/test_substrate_seam.py`).
+
+## Fresh-session quick answers
+
+- Is a claim true? → Spine store + evidence_refs, never commit messages.
+- Is construction allowed? → scaffold contract phase + LEGAL_TRANSITIONS.
+- How does a long wait work? → register on the W/R bridge, yield on the
+  FIFO event; watchdog is failure detection, never scheduling; the event
+  is a WAKE — authority is re-read after resume.
+- Who picks the provider? → the user layer, via ambient context;
+  construction never does (`test_provider_ownership.py` enforces).
+- What runs where? → CI on `rune-ci-desktop` runner; deploy stages to
+  `/mnt/workdrive/rune-code/<sha>`; data home pinned at
+  `/mnt/workdrive/rune` (RUNE_HOME); implementation worktree
+  `/mnt/workdrive/issue161-harness`.

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "rune"
-version = "0.66.58"
+version = "0.66.68"
 description = "Rune — a stateful coding assistant with persistent memory, orchestration, and TUI"
 requires-python = ">=3.10"
 

scripts/ci_event_wait.py

@@ -0,0 +1,262 @@
+#!/usr/bin/env python3
+"""ci_event_wait.py — T16 §16C/§16D: event-driven CI qualification wait.
+
+Replaces the deploy.yml sleep-30 → query → repeat polling loop with the
+EXISTING W/R event bridge (src/rune/event_wait.py — the SAME primitive the
+builder adapter uses; §16E: one bridge, adapters differ only in identity
+and authority probe).
+
+Flow (mandate §16C):
+
+    push / CI workflow starts
+        ↓
+    register correlation  (provider="gitea", event_type="ci_completed",
+                           run_ref="ci-<sha>", sha=<sha>)
+        ↓
+    WAITING_FOR_EVENT     (await_resume_event: select() on FIFO, no floor)
+        ↓
+    ci.yml final job POSTs /process-event  (the WAKE)
+        ↓
+    daemon matches → resume_cmd fires → FIFO token → controller resumes
+        ↓
+    canonical Gitea state INDEPENDENTLY re-read (deploy_qualify.py verdict)
+        ↓
+    verdict
+
+The event is a WAKE, never truth. The final verdict always comes from
+deploy_qualify.py (required-checks context + exact-SHA ci.yml run, both
+legs). Watchdog deadline is a FAILURE DETECTOR, never the scheduler:
+
+  - event arrives → authority re-read; pending → bounded re-arm until the
+    overall deadline (ack-window semantics, same as the daemon's wake sweep).
+  - event lost/suppressed → watchdog expires → ONE-SHOT authority query:
+    terminal verdict → recovered truthfully (no invented wait);
+    CI still running (non-terminal Gitea runs for the SHA) → re-arm;
+    nothing alive and no verdict → truthful failure (deployment refused).
+
+Exit codes: 0 = qualified (success), 1 = refused (failure/timeout).
+Output: one JSON line with the §16D latency-proof timestamps.
+"""
+from __future__ import annotations
+
+import json
+import os
+import subprocess
+import sys
+import time
+from pathlib import Path
+
+_SCRIPT_DIR = Path(__file__).resolve().parent
+sys.path.insert(0, str(_SCRIPT_DIR.parent / "src"))
+
+from rune import event_wait as ew  # noqa: E402
+
+REPO_ROOT = _SCRIPT_DIR.parent
+QUALIFY_SCRIPT = _SCRIPT_DIR / "deploy_qualify.py"
+
+
+def authority_verdict() -> str:
+    """ONE-SHOT canonical Gitea verdict via deploy_qualify.py — the only
+    completion truth. Never parses the event payload."""
+    proc = subprocess.run(
+        [sys.executable, str(QUALIFY_SCRIPT)],
+        capture_output=True, text=True, timeout=120,
+        env=dict(os.environ, PYTHONPATH=""),
+    )
+    out = (proc.stdout or "").strip().splitlines()
+    return (out[-1] if out else "pending").strip()
+
+
+def authority_after_wake(max_tries: int = 3, gap_secs: int = 4) -> str:
+    """Post-wake authority read. The daemon's own wake sweep uses a
+    bounded event-triggered retry (4s ×3) for the job-finished-microseconds-
+    ago ack window; a verdict read here can hit the same window. This is a
+    bounded schedule, not a polling loop — 3 reads max, then back to the
+    event wait / watchdog."""
+    verdict = "pending"
+    for attempt in range(max_tries):
+        verdict = authority_verdict()
+        if verdict != "pending":
+            return verdict
+        if attempt < max_tries - 1:
+            time.sleep(gap_secs)
+    return verdict
+
+
+def ci_alive() -> bool:
+    """Liveness probe for THIS adapter: any non-terminal Actions run for
+    the SHA in canonical Gitea (one-shot on watchdog expiry — not a poll).
+    Import the probe from deploy_qualify to keep ONE API client."""
+    sys.path.insert(0, str(_SCRIPT_DIR))
+    import deploy_qualify as dq  # noqa: E402  (spec-loaded scripts pattern)
+    import gitea_auth
+    server = os.environ.get("GITEA_SERVER", "").rstrip("/")
+    repo = os.environ.get("GITEA_REPO", "")
+    sha = os.environ.get("DEPLOY_SHA", "")
+    if not (server and repo and sha):
+        return False
+    token = gitea_auth.get_token() if gitea_auth.has_token() else ""
+    if not token:
+        return False
+    runs = dq._fetch_runs(server, repo, sha, token)
+    if not runs:
+        return False
+    terminal = {"success", "failure", "cancelled", "skipped"}
+    return any(str(r.get("status", "")).lower() not in terminal
+               for r in runs)
+
+
+def run(*, sha: str, deadline_secs: int,
+        watchdog_secs: int = 300,
+        daemon_url: str = ew.DEFAULT_DAEMON_URL) -> int:
+    op = f"ci-{sha}"
+    t0 = time.time()
+    # §16Y #10 race fix: the FIFO reader exists BEFORE registration, so a
+    # wake arriving in the registration window can always be delivered.
+    fifo_fd = ew.open_fifo_reader(op)
+    # Event replay (T16 §16C live finding): on a serialized runner the
+    # completion event usually arrives BEFORE the deploy job registers.
+    # A matching durable event means the wake already happened — skip
+    # the rendezvous entirely and go straight to the authority read.
+    prior = ew.find_matching_event("gitea", "ci_completed", op)
+    if prior is not None and prior.get("exit_code") is not None:
+        rec = ew.register_process_wait(
+            op, project="rune",
+            objective=f"event-driven CI qualification for {sha}",
+            watched_for="Gitea CI completion event (replayed from the "
+                        "durable event log)",
+            provider="gitea", event_type="ci_completed", sha=sha,
+            watchdog_secs=min(watchdog_secs, deadline_secs))
+        registered_at = rec["created_at"]
+        event_at = time.strftime(
+            "%Y-%m-%dT%H:%M:%SZ", time.gmtime())
+        resumed_at = event_at
+        # the event is a WAKE, never truth — canonical authority decides
+        verdict = authority_after_wake()
+        if verdict == "pending" and not ci_alive():
+            verdict = authority_verdict()
+        mode = ("event_replayed" if verdict != "pending"
+                else "authority_unresolved")
+        verdict_at = time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime())
+        out = {
+            "sha": sha,
+            "registered_at": registered_at,
+            "event_received_at": event_at,
+            "controller_resumed_at": resumed_at,
+            "verdict_at": verdict_at,
+            "mode": mode,
+            "rearm_reason": "completion event replayed from durable log "
+                            "(arrived before registration)",
+            "verdict": verdict,
+            "wall_secs": round(time.time() - t0, 3),
+        }
+        print(json.dumps(out))
+        return 0 if verdict == "success" else 1
+    rec = ew.register_process_wait(
+        op, project="rune",
+        objective=f"event-driven CI qualification for {sha}",
+        watched_for="Gitea CI completion event (wake; verdict re-read "
+                    "from canonical Gitea)",
+        provider="gitea", event_type="ci_completed", sha=sha,
+        watchdog_secs=min(watchdog_secs, deadline_secs))
+    registered_at = rec["created_at"]
+    deadline = time.monotonic() + deadline_secs
+    verdict = "pending"
+    mode = "event"
+    event_at = ""
+    resumed_at = ""
+    rearm_reason = ""
+    try:
+        while time.monotonic() < deadline:
+            remaining = max(1, int(deadline - time.monotonic()))
+            result = ew.await_resume_event(
+                op, watchdog_secs=min(remaining, watchdog_secs), pgid=None,
+                fifo_fd=fifo_fd)
+            if result["kind"] == "event":
+                event_at = event_at or time.strftime(
+                    "%Y-%m-%dT%H:%M:%SZ", time.gmtime())
+                resumed_at = time.strftime("%Y-%m-%dT%H:%M:%SZ",
+                                           time.gmtime())
+                # WAKE received — canonical authority decides, never the
+                # event payload
+                verdict = authority_after_wake()
+                if verdict != "pending":
+                    mode = "event"
+                    break
+                rearm_reason = "event seen; authority pending"
+                if not ci_alive():
+                    # CI terminal in Gitea but aggregator lagging: one
+                    # final truthful read, then stop
+                    verdict = authority_verdict()
+                    mode = ("event" if verdict != "pending"
+                            else "authority_unresolved")
+                    break
+                continue
+            if result["kind"] == "eof":
+                # writer gone without a token: the DURABLE record decides
+                rec = ew.operation_wait(op) or {}
+                if rec.get("status") in ("COMPLETED", "FAILED",
+                                         "READY_TO_RESUME", "RESUMING") or \
+                        (rec.get("status") == "RESUME_UNVERIFIED"
+                         and rec.get("process_event_evidence")):
+                    event_at = event_at or time.strftime(
+                        "%Y-%m-%dT%H:%M:%SZ", time.gmtime())
+                    resumed_at = time.strftime("%Y-%m-%dT%H:%M:%SZ",
+                                               time.gmtime())
+                    verdict = authority_after_wake()
+                    if verdict != "pending":
+                        mode = "event_recovered_from_durable_record"
+                        break
+                rearm_reason = "eof without token; durable record pending"
+                continue
+            # watchdog expiry: adapter liveness (Gitea runs) decides
+            # re-arm vs truthful recovery — failure detector, never
+            # scheduler
+            if ci_alive():
+                rearm_reason = "watchdog expiry; CI runs non-terminal " \
+                               "in Gitea — re-arm"
+                continue
+            verdict = authority_verdict()
+            if verdict == "pending":
+                mode = "watchdog_lost_event_no_verdict"
+            else:
+                mode = "watchdog_authority_recovered"
+            break
+    finally:
+        os.close(fifo_fd)
+    if verdict == "pending":
+        if mode == "event":
+            mode = "deadline_pending"
+        verdict = authority_verdict()  # final one-shot truth
+    verdict_at = time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime())
+    out = {
+        "sha": sha,
+        "registered_at": registered_at,
+        "event_received_at": event_at,
+        "controller_resumed_at": resumed_at,
+        "verdict_at": verdict_at,
+        "mode": mode,
+        "rearm_reason": rearm_reason,
+        "verdict": verdict,
+        "wall_secs": round(time.time() - t0, 3),
+    }
+    print(json.dumps(out))
+    return 0 if verdict == "success" else 1
+
+
+def main() -> int:
+    sha = os.environ.get("DEPLOY_SHA", "")
+    args = sys.argv[1:]
+    if "--sha" in args:
+        sha = args[args.index("--sha") + 1]
+    deadline = 900
+    if "--deadline" in args:
+        deadline = int(args[args.index("--deadline") + 1])
+    if not sha:
+        print(json.dumps({"error": "no sha (DEPLOY_SHA or --sha required)"}))
+        return 1
+    return run(sha=sha, deadline_secs=deadline)
+
+
+if __name__ == "__main__":
+    sys.exit(main())

scripts/notify_event_bridge.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python3
+"""notify_event_bridge.py — T16 §16C: ci.yml's final job posts the CI
+completion WAKE to the local W/R event bridge (daemon /process-event).
+
+The event is a WAKE, never completion truth: the waiting controller
+(ci_event_wait.py) re-reads the canonical Gitea verdict via
+deploy_qualify.py after resuming. Best-effort by design — if the bridge
+is unreachable the waiter's watchdog still recovers truthfully (that is
+the §16D lost-event contract), so a notify failure must NEVER fail CI.
+"""
+from __future__ import annotations
+
+import json
+import os
+import sys
+import time
+import urllib.request
+
+BRIDGE_URL = os.environ.get("RUNE_EVENT_BRIDGE_URL",
+                            "http://127.0.0.1:8791")
+
+
+def main() -> int:
+    sha = os.environ.get("CI_SHA", "")
+    result = os.environ.get("CI_RESULT", "")  # needs.<job>.result
+    if not sha:
+        print("notify-event-bridge: no CI_SHA — nothing to wake", flush=True)
+        return 0
+    payload = {
+        "operation_id": f"ci-{sha}",
+        "provider": "gitea",
+        "event_type": "ci_completed",
+        "sha": sha,
+        # exit_code is informational only; the wake is not the verdict
+        "exit_code": 0 if result == "success" else 1,
+        "completed_at": time.time(),
+    }
+    for attempt in range(2):
+        try:
+            req = urllib.request.Request(
+                BRIDGE_URL + "/process-event",
+                data=json.dumps(payload).encode(),
+                headers={"Content-Type": "application/json"})
+            with urllib.request.urlopen(req, timeout=5) as r:
+                resp = json.loads(r.read())
+            print(f"notify-event-bridge: woke bridge "
+                  f"(matched={resp.get('matched')})", flush=True)
+            return 0
+        except Exception as exc:  # noqa: BLE001 — wake is best-effort by design; failure is logged and the waiter's watchdog recovery covers a lost wake (§16D contract)
+            if attempt == 0:
+                time.sleep(2)
+            else:
+                print(f"notify-event-bridge: bridge unreachable ({exc}); "
+                      f"watchdog recovery covers a lost wake", flush=True)
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

scripts/research_event_relay.py

@@ -181,7 +181,8 @@ def run_stage_with_relay(tree, out_dir: str) -> dict:
     err_path = os.path.join(out_dir, f"stage_{stage}.mimo.err.log")
 
     mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-    mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+    from rune.provider_context import require_model
+    mimo_model = require_model()  # T16 §16F: provider layer owns selection
     mimo_agent = os.environ.get("MIMO_AGENT", "build")
 
     cmd = [

scripts/research_stage_executor.py

@@ -793,7 +793,8 @@ class StageExecutor:
         prompt = chr(10).join(pp)
 
         mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-        mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+        from rune.provider_context import require_model
+        mimo_model = require_model()  # T16 §16F: provider layer owns selection
         mimo_agent = os.environ.get("MIMO_AGENT", "build")
         events_path = os.path.join(self.artifact_dir, "stage_OUTWARD.events.jsonl")
         err_path = os.path.join(self.artifact_dir, "stage_OUTWARD.mimo.err.log")
@@ -933,7 +934,8 @@ class StageExecutor:
         prompt = chr(10).join(pp)
 
         mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-        mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+        from rune.provider_context import require_model
+        mimo_model = require_model()  # T16 §16F: provider layer owns selection
         mimo_agent = os.environ.get("MIMO_AGENT", "build")
         events_path = os.path.join(self.artifact_dir, "stage_APPLICABILITY.events.jsonl")
         err_path = os.path.join(self.artifact_dir, "stage_APPLICABILITY.mimo.err.log")
@@ -1137,7 +1139,8 @@ class StageExecutor:
         prompt = chr(10).join(pp)
 
         mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-        mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+        from rune.provider_context import require_model
+        mimo_model = require_model()  # T16 §16F: provider layer owns selection
         mimo_agent = os.environ.get("MIMO_AGENT", "build")
         events_path = os.path.join(self.artifact_dir, "stage_ADVERSARIAL.events.jsonl")
         err_path = os.path.join(self.artifact_dir, "stage_ADVERSARIAL.mimo.err.log")
@@ -1345,7 +1348,8 @@ class StageExecutor:
         prompt = chr(10).join(pp).replace("{n}", str(self.issue))
 
         mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-        mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+        from rune.provider_context import require_model
+        mimo_model = require_model()  # T16 §16F: provider layer owns selection
         mimo_agent = os.environ.get("MIMO_AGENT", "build")
         events_path = os.path.join(self.artifact_dir, "stage_INNOVATION.events.jsonl")
         err_path = os.path.join(self.artifact_dir, "stage_INNOVATION.mimo.err.log")
@@ -1534,7 +1538,8 @@ class StageExecutor:
         prompt = chr(10).join(pp)
 
         mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-        mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+        from rune.provider_context import require_model
+        mimo_model = require_model()  # T16 §16F: provider layer owns selection
         mimo_agent = os.environ.get("MIMO_AGENT", "build")
         events_path = os.path.join(self.artifact_dir, "stage_SYNTHESIS.events.jsonl")
         err_path = os.path.join(self.artifact_dir, "stage_SYNTHESIS.mimo.err.log")
@@ -1671,7 +1676,8 @@ class StageExecutor:
             prompt = chr(10).join(pp).replace("{n}", str(self.issue))
 
             mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-            mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+            from rune.provider_context import require_model
+            mimo_model = require_model()  # T16 §16F: provider layer owns selection
             mimo_agent = os.environ.get("MIMO_AGENT", "build")
             events_path = os.path.join(self.artifact_dir, "stage_QUALIFY.events.jsonl")
             err_path = os.path.join(self.artifact_dir, "stage_QUALIFY.mimo.err.log")

scripts/t14_builder_session.py

@@ -44,8 +44,18 @@ USER_REQUEST = (
     "list and complete practice items."
 )
 EXPECTED_SCOPE = ["scheduler.js", "scheduler.test.js", "package.json"]
-BUILDER_MODEL = os.environ.get("T14_BUILDER_MODEL",
-                                "Verboo Code/glm-5.3-flash")  # PL1 provider policy default
+# T16 §16F: NO silent default provider. The surrounding provider
+# layer sets T14_BUILDER_MODEL (falling back to the ambient MIMO_MODEL
+# context); construction never names a provider.
+import sys as _sys
+_sys.path.insert(0, str(__import__("pathlib").Path(__file__).resolve().parent.parent / "src"))
+from rune.provider_context import ProviderUnresolved, require_model  # noqa: E402
+try:
+    BUILDER_MODEL = os.environ.get("T14_BUILDER_MODEL", "").strip() \
+        or require_model()  # ambient provider context (MIMO_MODEL)
+except ProviderUnresolved as _exc:
+    _sys.stderr.write(str(_exc) + "\n")
+    raise SystemExit(3)  # 3 = PROVIDER_UNRESOLVED (no silent fallback)
 BUILDER_CMD = "mimo"
 
 
@@ -196,19 +206,42 @@ def _is_runtime_residue(path: str) -> bool:
 
 
 def audit_scope(project_dir: Path, before: dict,
-                allowed: list[str]) -> dict:
+                allowed: list[str], *,
+                runtime_declared: set[str] | None = None) -> dict:
+    """§15O PROVENANCE-AWARE residue classification.
+
+    The T14 prefix rule (.git/.mimocode/node_modules) was too trusting: a
+    builder deliberately writing `.mimocode/evil` would vanish from the
+    audit. Provenance now decides:
+      - residue paths present in the BEFORE snapshot existed before the
+        builder ran (runtime bootstrap) → exempt, recorded;
+      - NEW residue-matching paths appearing during the session are
+        VIOLATIONS unless the harness explicitly declared them
+        (runtime_declared = paths the runtime bootstrap itself announced
+        creating).
+    """
+    runtime_declared = runtime_declared or set()
     after = snapshot_files(project_dir)
     changed = sorted(set(after) - set(before))
     modified = sorted(k for k in after
                       if k in before and after[k] != before[k])
-    residue = sorted(f for f in changed + modified
-                     if _is_runtime_residue(f))
-    violations = sorted(f for f in changed + modified
-                        if f not in allowed and not _is_runtime_residue(f))
-    return {"created": [f for f in changed if not _is_runtime_residue(f)],
-            "modified": [f for f in modified if not _is_runtime_residue(f)],
-            "runtime_residue": residue,
-            "scope_violations": violations,
+    residue = []
+    violations = []
+    for f in changed + modified:
+        if f not in allowed:
+            if _is_runtime_residue(f):
+                if f in before or f in runtime_declared:
+                    residue.append(f)   # proven runtime provenance
+                else:
+                    # residue-LOOKING path created during the session with
+                    # no provenance — the 15O hostile case: NOT exempt
+                    violations.append(f)
+            else:
+                violations.append(f)
+    return {"created": [f for f in changed if f not in violations],
+            "modified": [f for f in modified if f not in violations],
+            "runtime_residue": sorted(set(residue)),
+            "scope_violations": sorted(set(violations)),
             "in_scope": not violations}
 
 
@@ -264,26 +297,14 @@ def complete_lifecycle(iface: ci.ConstructionInterface, contract_id: str,
 
 
 def discover_contract_id(store: Path) -> tuple[str, str] | None:
-    """Purely canonical discovery: walk the store's project scaffolds and
-    return (project, contract_id) for the most recent contract. No session
-    file, no transcript — exactly what a fresh builder may know."""
-    projects_dir = store / "projects"
-    if not projects_dir.exists():
+    """§15U: delegates to the canonical interface
+    (ConstructionInterface.discover_active_contract). This shim stays only
+    so existing driver call sites keep working."""
+    from rune import construction_interface as _ci
+    env = _ci.ConstructionInterface.discover_active_contract(str(store))
+    if not env.get("success"):
         return None
-    best: tuple[str, str, str] | None = None
-    for proj_current in sorted(projects_dir.glob("*/current.json")):
-        project = proj_current.parent.name
-        try:
-            doc = json.loads(proj_current.read_text())
-        except (OSError, ValueError):
-            continue
-        for e in doc["entities"]:
-            if e["kind"] == "construction_contract":
-                key = (e.get("observed_at") or "", project,
-                       e["id"].split(":", 1)[1])
-                if best is None or key > (best[0], best[1], best[2]):
-                    best = key
-    return (best[1], best[2]) if best else None
+    return env["ids"]["project"], env["ids"]["contract_id"]
 
 
 def seed_requirement_spec(store: Path, project: str) -> str:

scripts/t15_builder_session.py

@@ -0,0 +1,381 @@
+#!/usr/bin/env python3
+"""t15_builder_session.py — T15 §15Q: REAL event-driven builder proof.
+
+The difference from T14, enforced by construction:
+
+    launch → register WAITING_EVENT → YIELD (select on the bridge FIFO)
+    → builder exit event → resume immediately → reload canonical state
+    → readiness/exit canaries → independent verification → continuation
+
+NO sleep/poll loops anywhere on the normal path. The watchdog is only a
+failure detector (lost event / dead process), never the scheduler.
+
+Modes:
+    start     classify → activate → contract → real builder → event wait
+              → exit canary → verification → evidence-backed close
+    interrupt like start, but the builder process group is SIGKILLed at
+              the first source file → PROCESS_LOST → canonical state
+              persists for a fresh builder (run `resume` next)
+    resume    FRESH controller + fresh builder from canonical state only
+    status    print canonical construction state
+
+The builder is the real model process (mimo run — Verboo per the standing
+provider policy). Builder identity is recorded, never trusted.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import signal
+import subprocess
+import sys
+import time
+from pathlib import Path
+
+ROOT = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(ROOT / "src"))
+sys.path.insert(0, str(ROOT / "scripts"))
+
+import t14_builder_session as t14  # noqa: E402 — reuse the proven pieces
+from rune import construction_interface as ci  # noqa: E402
+from rune import event_wait as ew  # noqa: E402
+from rune import scaffold as sc  # noqa: E402
+from rune import structural_store as ss  # noqa: E402
+from rune.event_wait import DEFAULT_DAEMON_URL  # noqa: E402
+from rune.provider_context import ProviderUnresolved, require_model  # noqa: E402
+
+# T16 §16F: NO silent default provider — resolve from the ambient provider
+# context (T15_BUILDER_MODEL, falling back to MIMO_MODEL); loud failure
+# with exit code 3 when the provider layer has not selected one.
+try:
+    BUILDER_MODEL = os.environ.get("T15_BUILDER_MODEL", "").strip() \
+        or require_model()
+except ProviderUnresolved as _exc:
+    import sys as _sys
+    _sys.stderr.write(str(_exc) + "\n")
+    raise SystemExit(3)  # 3 = PROVIDER_UNRESOLVED (no silent fallback)
+DAEMON_URL = os.environ.get("T15_DAEMON_URL", DEFAULT_DAEMON_URL)
+
+
+def log(msg: str) -> None:
+    print(f"[t15] {msg}", flush=True)
+
+
+def readiness_canary(toolchain_hint: str) -> dict:
+    """§15F: may the build stage begin? Executable checks, not optimism.
+    Runtime checks derive from the verifier (15R generalisation)."""
+    checks = []
+    runtimes = (["node"] if "node" in VERIFIER else []) +                (["python3"] if "python" in VERIFIER else [])
+    for rt in runtimes:
+        r = subprocess.run([rt, "--version"], capture_output=True, text=True)
+        checks.append({"name": f"{rt}_present", "passed": r.returncode == 0,
+                       "evidence": (r.stdout or "").strip()[:30]})
+    r2 = subprocess.run(["mimo", "--version"], capture_output=True, text=True)
+    checks.append({"name": "builder_cli_present", "passed": r2.returncode == 0,
+                   "evidence": (r2.stdout or "").strip()[:40]})
+    try:
+        import urllib.request
+        with urllib.request.urlopen(DAEMON_URL + "/health", timeout=3) as resp:
+            ok = resp.status == 200
+        checks.append({"name": "event_bridge_healthy", "passed": ok,
+                       "evidence": DAEMON_URL})
+    except Exception as e:  # noqa: BLE001
+        checks.append({"name": "event_bridge_healthy", "passed": False,
+                       "evidence": str(e)[:80]})
+    verdict = "READY" if all(c["passed"] for c in checks) else "BLOCKED"
+    return {"canary": "readiness", "verdict": verdict, "checks": checks}
+
+
+# 15R: the driver is construction INFRASTRUCTURE — the request/scope are
+# parameters, never scheduler-specific constants. Defaults preserve the
+# proven fixture.
+REQUEST = os.environ.get(
+    "T15_REQUEST",
+    "Build me a small standalone practice scheduler that can create, "
+    "list and complete practice items.")
+SCOPE = [s for s in os.environ.get(
+    "T15_SCOPE", "scheduler.js,scheduler.test.js,package.json").split(",")
+    if s]
+PROJECT_NAME = os.environ.get("T15_PROJECT_NAME", "")
+SPEC_CONTRACT = os.environ.get("T15_SPEC_CONTRACT",
+                               "contract:practice-items-spec")
+TOOLCHAIN = os.environ.get("T15_TOOLCHAIN",
+                           "node >= 18 (node:test built-in runner, ES modules)")
+VERIFIER = os.environ.get("T15_VERIFIER", "node --test")
+
+
+def setup_construction(store: Path, project_dir: Path) -> tuple[str, str]:
+    """classify → activate → seed spec → contract → transitions. Reuses
+    the proven T14 canonical path unchanged."""
+    staged = t14.classify_and_activate(store, REQUEST)
+    project = staged["activation"]["project"]
+    assert staged["activation"]["mode"] == "PROJECT_BUILD"
+    _seed_spec(store, project)
+    iface = ci.ConstructionInterface(
+        builder=BUILDER_MODEL,
+        project_store=str(store / "projects" / project),
+        project_id=project)
+    sub = iface.submit_need(
+        REQUEST, origin="rune_native",
+        constraints={"environment": {"os": "linux",
+                                     "runtime": TOOLCHAIN,
+                                     "runner": VERIFIER},
+                     "contracts": [SPEC_CONTRACT]},
+        affected_surface=list(SCOPE),
+        evidence_refs=["t15-user-request", "node-v22-installed"],
+        approval_context="t15 user-approved fixture")
+    assert sub.get("success"), sub.get("errors")
+    contract_id = sub["ids"]["contract_id"]
+    for phase in ("approved", "building"):
+        tr = iface.transition(contract_id, phase, evidence_ref=f"t15 {phase}")
+        assert tr.get("success"), tr.get("errors")
+    log(f"contract {contract_id} → building")
+    return project, contract_id
+
+
+def _seed_spec(store: Path, project: str) -> str:
+    """Govern the requirement as a contract on the project surface
+    (generic — request/scope come from parameters)."""
+    proj = sc.Scaffold(str(store / "projects" / project), project)
+    doc = proj.load()
+    if ss.entity(doc, SPEC_CONTRACT) is None:
+        ss.add_entity(doc, id=SPEC_CONTRACT, kind="contract",
+                      identities={"objective": REQUEST},
+                      evidence_refs=["t15-request"])
+        ss.add_entity(doc, id="capability:planned-" + project[:24],
+                      kind="capability", maturity="DECLARED",
+                      evidence_refs=["t15-request"])
+        ss.add_relationship(doc, type="implements",
+                            from_id="capability:planned-" + project[:24],
+                            to_id=SPEC_CONTRACT, evidence_refs=["t15-request"])
+        proj.commit(doc, what="seed requirement spec (user request)")
+    return SPEC_CONTRACT
+
+
+def run_build_via_events(project_dir: Path, operation_id: str,
+                         projection: dict, contract_id: str,
+                         interrupt_after_first_file: bool) -> dict:
+    """THE 15Q core: real builder under the event notifier; the driver
+    YIELDS on the bridge FIFO. No sleep. Watchdog = failure detector."""
+    ew.register_process_wait(
+        operation_id, contract_id=contract_id,
+        project="t15-fixture", objective=REQUEST,
+        watched_for="builder process exit event", watchdog_secs=900)
+    # op marker rides the prompt (last argv) so cmdline liveness can find
+    # the build even if node regroups its children (T15Q live finding)
+    builder_cmd = ["mimo", "run", "-m", BUILDER_MODEL, "--format", "json",
+                   t14.builder_prompt(projection, resumed=False)
+                   + f" [op:{operation_id}]"]
+    notifier = ew.run_builder_under_events(builder_cmd, operation_id,
+                                           daemon_url=DAEMON_URL)
+    log(f"builder launched (pgid {notifier.pid}); driver YIELDS on the "
+        f"event bridge — no polling")
+    if interrupt_after_first_file:
+        # kill at first source file: the interruption is real, the event
+        # never arrives (the notifier dies with the group) — the watchdog
+        # must classify PROCESS_LOST, never silently wait.
+        # TEST_TIMING: this 0.5s poll exists ONLY on the deliberate
+        # interrupt path (killing at a checkpoint), never on the normal
+        # continuation path.
+        while True:
+            files = [p for p in project_dir.iterdir()
+                     if p.suffix in (".mjs", ".js", ".json")]
+            if files:
+                break
+            time.sleep(0.5)
+        log("INTERRUPTING builder (first file landed) — SIGKILL group")
+        os.killpg(notifier.pid, signal.SIGKILL)
+    result = ew.await_resume_event(operation_id, watchdog_secs=900,
+                                   pgid=notifier.pid)
+    return result
+
+
+def finish_via_canaries(store: Path, project: str, contract_id: str,
+                        project_dir: Path, before: dict,
+                        resume_note: str) -> int:
+    """Reload canonical state → scope audit → INDEPENDENT verification
+    (15G test truth) → evidence-backed lifecycle."""
+    iface = ci.ConstructionInterface(
+        builder=BUILDER_MODEL, project_store=str(store / "projects" / project),
+        project_id=project)
+    ctx = iface.construction_context(contract_id)
+    assert ctx["state"]["active"] or ctx["state"]["phase"] == "closed"
+    log(f"canonical state: phase={ctx['state']['phase']} "
+        f"scope={ctx['state']['scope']}")
+    # exit canary: scope audit with provenance-aware residue (15O)
+    audit = t14.audit_scope(project_dir, before, list(SCOPE))
+    # 15G: verification truth — node --test must actually run tests
+    # 15G test truth via the CONTRACT's verifier (parameterized — 15R
+    # generalisation: python unittest/pytest, node --test, etc.)
+    truth = subprocess.run(
+        ["bash", "-c", VERIFIER], cwd=str(project_dir),
+        capture_output=True, text=True, timeout=300)
+    import re
+    count = None
+    for pat in (r"# pass (\d+)",       # node --test TAP summary
+                r"(\d+) passed",        # pytest
+                r"Ran (\d+) tests"):    # python unittest
+        m = re.search(pat, (truth.stdout or "") + (truth.stderr or ""))
+        if m:
+            count = int(m.group(1))
+            break
+    test_count = count if count is not None else 0
+    log(f"verification: exit={truth.returncode} tests={test_count}")
+    if test_count == 0:
+        log("FALSE GREEN REFUSED: zero tests executed (15G)")
+        return 1
+    if audit["scope_violations"]:
+        log(f"closure REFUSED: scope violations {audit['scope_violations']}")
+        return 1
+    builder_status = "UNKNOWN"
+    result = t14.complete_lifecycle(iface, contract_id, audit,
+                                    {"cmd": "node --test",
+                                     "exit_code": truth.returncode},
+                                    builder_status)
+    summary = {
+        "event_driven": True,
+        "builder_model": BUILDER_MODEL,
+        "resume_note": resume_note,
+        "scope_audit": audit,
+        "verification": {"exit_code": truth.returncode,
+                         "test_count": test_count},
+        "lifecycle": result,
+    }
+    print(json.dumps(summary, indent=1))
+    return 0 if result.get("closed") else 1
+
+
+def discover(store: Path) -> tuple[str, str] | None:
+    """§15U: delegates to the canonical interface discovery (via the t14
+    shim — one hop, no duplication)."""
+    return t14.discover_contract_id(store)
+
+
+def cmd_start(store: Path, project_dir: Path, interrupt: bool) -> int:
+    ready = readiness_canary("node")
+    assert ready["verdict"] == "READY", f"readiness BLOCKED: {ready['checks']}"
+    log(f"readiness READY ({len(ready['checks'])} checks)")
+    before = t14.snapshot_files(project_dir)
+    project, contract_id = setup_construction(store, project_dir)
+    operation_id = f"t15-{contract_id}"
+    iface = ci.ConstructionInterface(
+        builder=BUILDER_MODEL, project_store=str(store / "projects" / project),
+        project_id=project)
+    projection = t14.build_projection(iface, contract_id,
+                                      phase_note="initial build")
+    result = run_build_via_events(project_dir, operation_id, projection,
+                                  contract_id, interrupt)
+    if result["kind"] == "watchdog":
+        # §15E canonical-state-wins: process-table absence is NOT truth
+        # (detached builders re-parent their workers — T15Q live finding).
+        # The durable waitstore decides: a recorded verdict means the
+        # builder completed and the event WAS processed.
+        rec = ew.resume_or_recover(operation_id)
+        if rec["kind"] == "event":
+            log(f"watchdog fired but canonical state shows the build "
+                f"completed (verdict={rec['wait'].get('verdict')}) — "
+                f"continuing from authority")
+            return finish_via_canaries(store, project, contract_id,
+                                       project_dir, before,
+                                       "watchdog→canonical recovery")
+        log("PROCESS_LOST: no event, no durable verdict — fresh builder "
+            "resumes from authority only")
+        return 3
+    wait = result.get("wait", {})
+    log(f"event-driven resume: verdict={wait.get('verdict')} "
+        f"exit={wait.get('process_event_evidence', {}).get('exit_code')}")
+    return finish_via_canaries(store, project, contract_id, project_dir,
+                               before, "event-driven completion")
+
+
+def cmd_resume(store: Path, project_dir: Path) -> int:
+    found = discover(store)
+    assert found, "no contract in canonical state"
+    project, contract_id = found
+    log(f"FRESH builder resume: {project}/{contract_id} (canonical only)")
+    before = t14.snapshot_files(project_dir)
+    iface = ci.ConstructionInterface(
+        builder=BUILDER_MODEL + "-resumed",
+        project_store=str(store / "projects" / project),
+        project_id=project)
+    projection = t14.build_projection(
+        iface, contract_id, phase_note="resumed after real interruption")
+    operation_id = f"t15-resume-{contract_id}"
+    ew.register_process_wait(
+        operation_id, contract_id=contract_id, project=project,
+        objective=REQUEST, watched_for="builder exit event",
+        watchdog_secs=900)
+    builder_cmd = ["mimo", "run", "-m", BUILDER_MODEL, "--format", "json",
+                   t14.builder_prompt(projection, resumed=True)
+                   + f" [op:{operation_id}]"]
+    notifier = ew.run_builder_under_events(builder_cmd, operation_id,
+                                           daemon_url=DAEMON_URL)
+    log("driver YIELDS on the bridge — no polling")
+    result = ew.await_resume_event(operation_id, watchdog_secs=900,
+                                   pgid=notifier.pid)
+    if result["kind"] == "watchdog":
+        rec = ew.resume_or_recover(operation_id)
+        if rec["kind"] == "event":
+            log(f"watchdog fired but canonical state shows the resume "
+                f"build completed (verdict={rec['wait'].get('verdict')})")
+            return finish_via_canaries(store, project, contract_id,
+                                       project_dir, before,
+                                       "resume via canonical recovery")
+        log("PROCESS_LOST on resume — UNKNOWN_INTERRUPTION")
+        return 4
+    return finish_via_canaries(store, project, contract_id, project_dir,
+                               before, "fresh-builder event-driven resume")
+
+
+def cmd_finish(store: Path, project_dir: Path) -> int:
+    """Complete lifecycle from CANONICAL state only (waitstore verdict +
+    filesystem reality). Used after a watchdog→canonical recovery."""
+    found = discover(store)
+    assert found, "no contract in canonical state"
+    project, contract_id = found
+    before = t14.snapshot_files(project_dir)
+    return finish_via_canaries(store, project, contract_id, project_dir,
+                               before, "canonical-only completion")
+
+
+def cmd_status(store: Path) -> int:
+    found = discover(store)
+    if not found:
+        print(json.dumps({"active": False}))
+        return 0
+    project, contract_id = found
+    iface = ci.ConstructionInterface(
+        builder="inspector", project_store=str(store / "projects" / project),
+        project_id=project)
+    print(json.dumps(iface.construction_context(contract_id)["state"],
+                     indent=1, default=str))
+    return 0
+
+
+def main() -> int:
+    import argparse
+    ap = argparse.ArgumentParser(description=__doc__)
+    ap.add_argument("mode", choices=["start", "resume", "finish",
+                                     "status"])
+    ap.add_argument("--store", required=True)
+    ap.add_argument("--project-dir", required=False)
+    ap.add_argument("--interrupt-after-first-file", action="store_true")
+    args = ap.parse_args()
+    store = Path(args.store).resolve()
+    project_dir = (Path(args.project_dir).resolve()
+                   if args.project_dir else None)
+    if project_dir:
+        project_dir.mkdir(parents=True, exist_ok=True)
+    if args.mode == "start":
+        return cmd_start(store, project_dir,
+                         args.interrupt_after_first_file)
+    if args.mode == "resume":
+        return cmd_resume(store, project_dir)
+    if args.mode == "finish":
+        return cmd_finish(store, project_dir)
+    return cmd_status(store)
+
+
+if __name__ == "__main__":
+    sys.exit(main())

scripts/type_check_ratchet.py

@@ -31,7 +31,21 @@ def main() -> int:
     # PYTHONNOUSERSITE=1 pins both contexts to the same effective
     # environment (verified: identical 117-error arithmetic). Local
     # preflight now predicts CI instead of diverging from it.
-    env = dict(os.environ, PYTHONNOUSERSITE="1")
+    # §16U ROOT-CAUSE COMPLETION (T16): PYTHONNOUSERSITE alone is NOT
+    # sufficient when a DIFFERENT venv is active (e.g. an ad-hoc CI venv):
+    # the pyright wrapper resolves the ambient VIRTUAL_ENV/PATH
+    # interpreter and analyzes against ITS site-packages — reproduced
+    # live (117 vs 124 on the same tree, driven by venv-activation state
+    # alone). The CI runner resolves the AMBIENT python (no venv is
+    # activated in the type-check job), so the ratchet neutralizes every
+    # ambient-venv signal before invoking pyright — analysis context is
+    # now identical to CI regardless of the calling shell.
+    env = {k: v for k, v in os.environ.items()
+           if k not in ("VIRTUAL_ENV",)}
+    env["PYTHONNOUSERSITE"] = "1"
+    env["PATH"] = os.pathsep.join(
+        p for p in env.get("PATH", "").split(os.pathsep)
+        if "ci-venv" not in p and ".type-venv" not in p and "venv/bin" not in p)
     result = subprocess.run(
         cmd + [str(ROOT / "src" / "rune"), "--outputjson"],
         capture_output=True,

src/rune/__init__.py

@@ -6,7 +6,7 @@ This is the main entry point for the Rune program.
 import json
 from pathlib import Path
 
-__version__ = "0.66.58"
+__version__ = "0.66.68"
 
 # One build identity (handoff 2026-09-10 §4): the semantic version above is
 # the repo-level truth (auto-bumped at merge); the DEPLOYED revision (exact

src/rune/canary_gitea_automation.py

@@ -231,7 +231,7 @@ def main() -> int:
     )
     check(
         "Rune minor version advertises provisioning and restore verification",
-        '__version__ = "0.66.58"' in rune_version,
+        '__version__ = "0.66.68"' in rune_version,
         failures,
     )
     # #169 policy update: auto-bump on merge is sanctioned (Nick directive:

src/rune/construction_commands.py

@@ -0,0 +1,173 @@
+"""construction_commands.py — T16 §16J: bounded validated inbound
+command interface for future Substrate integration.
+
+The future Substrate ASKS; Scaffolding DECIDES. Every command:
+
+  - carries identity (command_id, correlation_id, construction_id)
+  - is validated by Scaffolding against the existing lifecycle
+    (LEGAL_TRANSITIONS — no new lifecycle invented)
+  - cannot directly mutate storage (commands return validated intents;
+    the Construction Interface executes them through its own governed
+    path — the caller never writes the store)
+  - is idempotent where applicable (repeat delivery of the same
+    command_id returns the same recorded outcome)
+
+Only commands that fit EXISTING semantics safely are exposed. PAUSE and
+CANCEL are intentionally NOT exposed: the lifecycle has no paused phase
+and abandonment needs user provenance — a future Substrate asking for
+them gets a refusal, not a half-measure.
+"""
+from __future__ import annotations
+
+import time
+import uuid
+
+from rune.scaffold import LEGAL_TRANSITIONS
+
+COMMANDS = (
+    "REQUEST_STATE", "START", "RESUME", "ACKNOWLEDGE_BLOCKER",
+    "PROVIDE_APPROVAL", "PROVIDE_CONTEXT",
+)
+
+# Command → the interface operation it asks for (the interface still
+# decides: validation failures return refused, never raise into storage).
+_COMMAND_TO_OP = {
+    "REQUEST_STATE": "resume",       # read-only state projection
+    "START": "transition",           # draft → approved (then building)
+    "RESUME": "resume",              # re-read canonical state
+    "ACKNOWLEDGE_BLOCKER": "transition",  # blocked → building (retry path)
+    "PROVIDE_APPROVAL": "transition",     # draft → approved with provenance
+    "PROVIDE_CONTEXT": "amend_scope",     # context arrives as scope amendment
+}
+
+# In-memory idempotency ledger: command_id → recorded outcome. The
+# Scaffolding lifecycle is the authority; this ledger only makes repeat
+# delivery return the same answer (no double transition).
+_outcomes: dict[str, dict] = {}
+
+
+def command(command_type: str, *, construction_id: str,
+            correlation_id: str = "", command_id: str = "",
+            payload: dict | None = None,
+            scaffold=None) -> dict:
+    """Validate an inbound command. Returns a validated intent envelope:
+
+      {"accepted": True, "op": <interface op>, "args": {...},
+       "command_id": ..., "idempotent_replay": bool}
+
+    or {"accepted": False, "reason": ...}. The caller executes the intent
+    through the Construction Interface — it never touches storage."""
+    payload = payload or {}
+    command_id = command_id or f"cmd-{uuid.uuid4().hex[:8]}"
+    if command_id in _outcomes:
+        prior = dict(_outcomes[command_id])
+        prior["idempotent_replay"] = True
+        return prior
+    if command_type not in COMMANDS:
+        return _refuse(command_id, construction_id, correlation_id,
+                       f"unknown command {command_type!r} "
+                       f"(supported: {', '.join(COMMANDS)})")
+    if not construction_id:
+        return _refuse(command_id, construction_id, correlation_id,
+                       "construction_id is required")
+    intent = _validate(command_type, construction_id,
+                       correlation_id, command_id, payload, scaffold)
+    _outcomes[command_id] = intent
+    return intent
+
+
+def _refuse(command_id: str, construction_id: str,
+            correlation_id: str, reason: str) -> dict:
+    return {"accepted": False, "reason": reason,
+            "command_id": command_id, "construction_id": construction_id,
+            "correlation_id": correlation_id,
+            "recorded_at": time.strftime("%Y-%m-%dT%H:%M:%SZ",
+                                         time.gmtime()),
+            "idempotent_replay": False}
+
+
+def _validate(command_type: str, construction_id: str,
+              correlation_id: str, command_id: str,
+              payload: dict, scaffold) -> dict:
+    """Validate against the CURRENT canonical state — the lifecycle
+    decides, not the requester. The scaffold (canonical store handle) is
+    supplied by the serving layer; the client only ever sees the returned
+    envelope, never the store."""
+    if scaffold is None:
+        return _refuse(command_id, construction_id, correlation_id,
+                       "no serving scaffold bound (server-side wiring)")
+    doc = scaffold.load()
+    contract = None
+    for e in doc.get("entities", []):
+        if e.get("id") == f"contract:{construction_id}":
+            contract = e
+            break
+    if contract is None:
+        return _refuse(command_id, construction_id, correlation_id,
+                       f"unknown construction {construction_id!r}")
+    phase = contract.get("phase", "")
+
+    if command_type == "REQUEST_STATE":
+        return _accept(command_id, construction_id, correlation_id,
+                       "resume", {"contract_id": construction_id})
+    if command_type == "RESUME":
+        if phase in ("closed", "abandoned"):
+            return _refuse(command_id, construction_id, correlation_id,
+                           f"construction is {phase} — resume is invalid")
+        return _accept(command_id, construction_id, correlation_id,
+                       "resume", {"contract_id": construction_id})
+    if command_type == "START":
+        if phase != "draft":
+            return _refuse(command_id, construction_id, correlation_id,
+                           f"START requires draft phase (current: {phase})")
+        return _accept(command_id, construction_id, correlation_id,
+                       "transition",
+                       {"contract_id": construction_id,
+                        "phase": "approved"})
+    if command_type == "PROVIDE_APPROVAL":
+        if phase != "draft":
+            return _refuse(command_id, construction_id, correlation_id,
+                           f"approval applies to draft (current: {phase})")
+        if payload.get("approved_by") != "user":
+            # lifecycle rule: approval provenance must be the user —
+            # Substrate cannot approve on the user's behalf
+            return _refuse(command_id, construction_id, correlation_id,
+                           "approval provenance must be user "
+                           "(Substrate cannot approve for the user)")
+        return _accept(command_id, construction_id, correlation_id,
+                       "transition",
+                       {"contract_id": construction_id,
+                        "phase": "approved",
+                        "approved_by": "user"})
+    if command_type == "ACKNOWLEDGE_BLOCKER":
+        if phase not in ("failed", "building", "verifying"):
+            return _refuse(command_id, construction_id, correlation_id,
+                           f"no blockable phase (current: {phase})")
+        return _accept(command_id, construction_id, correlation_id,
+                       "transition",
+                       {"contract_id": construction_id,
+                        "phase": "building"})
+    if command_type == "PROVIDE_CONTEXT":
+        paths = payload.get("add_paths") or []
+        if not paths:
+            return _refuse(command_id, construction_id, correlation_id,
+                           "PROVIDE_CONTEXT requires add_paths")
+        return _accept(command_id, construction_id, correlation_id,
+                       "amend_scope",
+                       {"contract_id": construction_id,
+                        "add_paths": paths,
+                        "reason": payload.get("reason", "substrate context"),
+                        "authority": "user"
+                        if payload.get("approved_by") == "user" else "substrate"})
+    return _refuse(command_id, construction_id, correlation_id,
+                   "unreachable validator branch")
+
+
+def _accept(command_id: str, construction_id: str, correlation_id: str,
+            op: str, args: dict) -> dict:
+    return {"accepted": True, "op": op, "args": args,
+            "command_id": command_id, "construction_id": construction_id,
+            "correlation_id": correlation_id,
+            "recorded_at": time.strftime("%Y-%m-%dT%H:%M:%SZ",
+                                         time.gmtime()),
+            "idempotent_replay": False}

src/rune/construction_interface.py

@@ -214,7 +214,11 @@ class ConstructionInterface:
         """§28 schema negotiation: builders detect the supported interface
         schema and operation set BEFORE calling operations. Also exposes the
         error-state vocabulary so builders can switch on states without
-        parsing prose. Read-only; no contract required."""
+        parsing prose. Read-only; no contract required.
+        §15N: full argument schemas included — a fresh builder discovers
+        required types/fields deterministically instead of trial-and-error
+        (the T14 report_observation dict-vs-list + missing-entity mistakes
+        are exactly what this prevents)."""
         return _envelope(
             "capabilities",
             success=True,
@@ -223,6 +227,7 @@ class ConstructionInterface:
                 "mode": self.mode,
                 "project": self.scaffold.project_id,
                 "operations": [
+                    "discover_active_contract",
                     "submit_need",
                     "inspect_decision",
                     "get_scope",
@@ -236,6 +241,67 @@ class ConstructionInterface:
                     "capabilities",
                 ],
                 "error_states": sorted(v for k, v in vars(Errors).items() if not k.startswith("_")),
+                # ── §15N argument schemas (discovery over guessing) ──
+                "argument_schemas": {
+                    "submit_need": {
+                        "objective": {"type": "str", "required": True},
+                        "origin": {"type": "str", "default": "builder"},
+                        "constraints": {"type": "dict", "required": False,
+                                        "notes": ("{'environment': {...}, "
+                                                  "'contracts': [...]} — "
+                                                  "environment missing "
+                                                  "degrades to "
+                                                  "INSUFFICIENT_EVIDENCE")},
+                        "affected_surface": {"type": "list[str]",
+                                             "required": False},
+                        "evidence_refs": {"type": "list[str]",
+                                          "required": True,
+                                          "notes": ("empty list degrades "
+                                                    "fit resolution")},
+                        "approval_context": {"type": "str",
+                                             "required": False},
+                    },
+                    "report_observation": {
+                        "observations": {
+                            "type": "list[dict]",
+                            "required": True,
+                            "item_required_fields": ["entity"],
+                            "item_optional_fields": ["source", "detail"],
+                            "notes": ("a LIST of dicts, each with an "
+                                      "'entity' key — a bare dict or a "
+                                      "dict without 'entity' is rejected "
+                                      "(NEED_INCOMPLETE)"),
+                        },
+                    },
+                    "verify": {
+                        "changed_files": {"type": "list[str] | None",
+                                          "required": False,
+                                          "notes": ("scope-checked against "
+                                                    "the contract scope; "
+                                                    "out-of-scope files → "
+                                                    "SCOPE_VIOLATION")},
+                    },
+                    "transition": {
+                        "phase": {"type": "str", "required": True,
+                                  "allowed": list(sc.PHASES)},
+                    },
+                    "close": {
+                        "observed_evidence": {
+                            "type": "list[dict]", "required": True,
+                            "notes": ("evidence INPUT only — closure is "
+                                      "decided by verification truth, "
+                                      "never by builder claim")},
+                        "scope_files": {"type": "list[str] | None",
+                                        "required": False},
+                    },
+                },
+                "lifecycle": {
+                    "phases": list(sc.PHASES),
+                    "legal_transitions": {k: sorted(v) for k, v in
+                                          sc.LEGAL_TRANSITIONS.items()},
+                    "scope_amendment": ("user-gated via amend_scope — "
+                                        "builders cannot broaden scope"),
+                },
             },
         )
 
@@ -617,6 +683,53 @@ class ConstructionInterface:
             },
         )
 
+    @staticmethod
+    def discover_active_contract(store_path: str) -> dict:
+        """§15U canonical resume discovery: walk the store's project
+        scaffolds AND the main store's own current.json, returning the
+        most recent contract's (project, contract_id). Migrated here from
+        the T14/T15 session drivers — a fresh builder with NO transcript
+        discovers what to resume through the INTERFACE, not a script.
+        Pure readback: no session file, no transcript, no mutation."""
+        from pathlib import Path as _P
+        import json as _json
+        root = _P(store_path)
+        candidates: list[tuple] = []
+        for proj_current in sorted(root.glob("projects/*/current.json")):
+            # per-project scaffold: (file, project_id)
+            candidates.append((proj_current, proj_current.parent.name))
+        main = root / "current.json"
+        if main.exists():
+            # main store: contracts carry their project in identities
+            candidates.append((main, None))
+        if not candidates:
+            return _envelope("discover_active_contract", success=False,
+                             errors=[{"error": Errors.UNKNOWN_CONTRACT,
+                                      "detail": "no projects in store"}])
+        best = None
+        best_project = ""
+        for proj_current, project in candidates:
+            try:
+                doc = _json.loads(proj_current.read_text())
+            except (OSError, ValueError):
+                continue
+            for e in doc["entities"]:
+                if e["kind"] == "construction_contract":
+                    proj = project or e.get("identities", {}).get(
+                        "project", root.name)
+                    key = (e.get("observed_at") or "", proj,
+                           e["id"].split(":", 1)[1])
+                    if best is None or key > (best[0], best[1], best[2]):
+                        best = key
+                        best_project = proj
+        if best is None:
+            return _envelope("discover_active_contract", success=False,
+                             errors=[{"error": Errors.UNKNOWN_CONTRACT,
+                                      "detail": "no contracts in store"}])
+        return _envelope("discover_active_contract", success=True,
+                         ids={"project": best_project,
+                              "contract_id": best[2]})
+
     def resume(self, contract_id: str) -> dict:
         doc = self.scaffold.load()
         c = self.scaffold.contract(doc, contract_id)

src/rune/construction_mode.py

@@ -64,6 +64,10 @@ _STANDALONE_WORDS = {
     "scaffold a new",
     "new web",
 }
+# T15 §15J: extensionless filenames that bind a scoped target when
+# mentioned explicitly (dotted/path tokens go through classify_surface)
+_KNOWN_FILES = {"readme", "makefile", "dockerfile", "changelog", "license",
+                "gemfile", "procfile", "jenkinsfile", "justfile"}
 _LOCAL_WORDS = {
     "typo",
     "small fix",
@@ -267,6 +271,27 @@ def classify_mode(
     standalone = _contains_any(obj.lower(), _STANDALONE_WORDS)
     local = _contains_any(obj.lower(), _LOCAL_WORDS)
     scoped = _contains_any(obj.lower(), _SCOPING_WORDS)
+    # T15 §15J: an explicit FILE mention binds a scoped target ("fix the
+    # typo in the README", "update scheduler.js"). Deterministic lexical
+    # evidence in the same corroboration style as the other lexicons;
+    # dotted/path tokens delegated to the ONE surface authority
+    # (construction_decision.classify_surface) — lazily imported (cycle:
+    # construction_decision imports this module's classify_mode).
+    file_scoped = False
+    for w in re.split(r"[^a-z0-9_./-]+", obj.lower()):
+        if not w or len(w) < 3:
+            continue
+        if w in _KNOWN_FILES or "." in w or "/" in w:
+            try:
+                from rune.construction_decision import classify_surface
+                if classify_surface(w) in ("source_path", "workflow_config"):
+                    file_scoped = True
+                    break
+            except Exception:  # noqa: BLE001,S110 — authority unavailable:
+                pass           # lexical fallback decides; never fatal
+            if w in _KNOWN_FILES:
+                file_scoped = True
+                break
 
     # 3. PROJECT_BUILD: standalone/new-lifecycle language AND no existing
     #    project for the target (project existence materially matters, §5)
@@ -286,10 +311,12 @@ def classify_mode(
             mode, conf = "PROJECT_BUILD", 0.9
             bootstrap = True
     # 4. LOCAL_CHANGE: explicit smallness + scoped target + no standalone
-    elif local and (scoped or target_project or project_exists):
+    elif local and (scoped or file_scoped or target_project or project_exists):
         ev.append(f"smallness markers: {local[:3]}")
         if scoped:
             ev.append(f"scoping words bind to known target: {scoped[:3]}")
+        if file_scoped:
+            ev.append("file mention binds scoped target")
         mode, conf = "LOCAL_CHANGE", 0.8
     # 5. EXISTING_PROJECT_CHANGE: scoping to a known/existing target.
     #    T11 §1: a change scoped to a project with PROVEN non-existence must

src/rune/construction_projection.py

@@ -0,0 +1,135 @@
+"""construction_projection.py — T16 §16I: stable versioned PUBLIC
+projection of construction state for future Substrate integration.
+
+This is a PROJECTION, never a second authority. All fields are derived
+(read-only) from the canonical scaffold store (Scaffold.load) plus the
+live wait bridge state (event_wait.operation_wait). A consumer (the
+future Substrate, the mock client in tests) must know NOTHING about
+current.json layout or internal entity arrays — only this schema.
+
+Schema versioning: SCHEMA_VERSION is bumped on any field change; consumers
+negotiate by reading schema_version first. Unknown future fields are
+ignored by readers (forward-tolerant); removed fields bump the major.
+"""
+from __future__ import annotations
+
+from typing import Any
+
+SCHEMA_VERSION = 1
+
+# The mandate's required projection fields (§16I). None may carry live
+# provider/account secrets or internal storage layout.
+FIELDS = (
+    "schema_version", "project_id", "originating_goal_id",
+    "construction_id", "contract_id",
+    "mode", "depth", "phase", "stage",
+    "operation_id", "waiting_state", "correlation_id",
+    "builder_state", "blockers", "approval_required",
+    "readiness_canary", "liveness_canary", "exit_canary",
+    "verification_state", "checkpoint", "evidence_refs",
+    "failure_class", "completion_state", "candidate_next_action",
+)
+
+
+def _contract_entity(doc: dict, contract_id: str) -> dict | None:
+    for e in doc.get("entities", []):
+        if e.get("id") == f"contract:{contract_id}":
+            return e
+    return None
+
+
+def project(
+    scaffold, contract_id: str, *,
+    originating_goal_id: str = "",
+    operation_id: str = "",
+    correlation_id: str = "",
+    stage: str = "",
+    builder_state: str = "",
+    verification_state: str = "",
+    checkpoint: str = "",
+    failure_class: str = "",
+    completion_state: str = "",
+) -> dict:
+    """Build the stable public projection for one contract. `scaffold` is
+    a Rune Scaffold instance (the canonical store); all live fields are
+    derived from it. Optional live fields (waiting_state etc.) are filled
+    from the event bridge when operation_id is supplied."""
+    doc = scaffold.load()
+    c = _contract_entity(doc, contract_id) or {}
+    ids = c.get("identities", {})
+    ext = c.get("ext", {})
+    trace = ext.get("decision_trace", {})
+
+    waiting_state = ""
+    live_correlation = correlation_id
+    if operation_id:
+        try:
+            from rune import event_wait as ew
+            rec = ew.operation_wait(operation_id)
+            if rec is not None:
+                waiting_state = rec.get("status", "")
+                live_correlation = correlation_id or rec.get("wait_id", "")
+        except OSError:  # bridge data root unreadable → degrade, never lie
+            waiting_state = "BRIDGE_UNREACHABLE"
+        except RuntimeError:  # waitstore not resolvable in this context
+            waiting_state = "BRIDGE_UNREACHABLE"
+
+    checkpoints = c.get("checkpoints", [])
+    evidence_refs = []
+    for cp in checkpoints:
+        ev = cp.get("evidence", "")
+        if ev:
+            evidence_refs.append(ev[:120])
+
+    approval = trace.get("approval", {})
+    approval_required = bool(
+        approval.get("required") or approval.get("state") == "REQUIRED"
+        or c.get("phase") == "awaiting_approval")
+
+    blockers: list[str] = []
+    blk = ids.get("blocker")
+    if isinstance(blk, dict):
+        blockers.append(f"{blk.get('class', 'BLOCKED')}: "
+                        f"{blk.get('evidence', '')[:120]}")
+    elif isinstance(blk, str) and blk:
+        blockers.append(blk)
+
+    return {
+        "schema_version": SCHEMA_VERSION,
+        "project_id": getattr(scaffold, "project_id", ""),
+        "originating_goal_id": originating_goal_id,
+        "construction_id": contract_id,
+        "contract_id": contract_id,
+        "mode": ids.get("construction_mode", "") or "",
+        "depth": ids.get("depth", "") or "",
+        "phase": c.get("phase", ""),
+        "stage": stage,
+        "operation_id": operation_id,
+        "waiting_state": waiting_state,
+        "correlation_id": live_correlation,
+        "builder_state": builder_state,
+        "blockers": blockers,
+        "approval_required": approval_required,
+        "readiness_canary": ext.get("readiness_canary", ""),
+        "liveness_canary": ext.get("liveness_canary", ""),
+        "exit_canary": ext.get("exit_canary", ""),
+        "verification_state": verification_state,
+        "checkpoint": checkpoint,
+        "evidence_refs": evidence_refs,
+        "failure_class": failure_class,
+        "completion_state": completion_state,
+        "candidate_next_action": ids.get("next_action", ""),
+    }
+
+
+def validate(projection: dict) -> list[str]:
+    """Check a projection against the schema. Returns a list of errors
+    (empty = valid). Consumers use this before trusting a projection."""
+    errors = []
+    if projection.get("schema_version") != SCHEMA_VERSION:
+        errors.append(f"schema_version: expected {SCHEMA_VERSION}, got "
+                      f"{projection.get('schema_version')!r}")
+    for f in FIELDS:
+        if f not in projection:
+            errors.append(f"missing field: {f}")
+    return errors

src/rune/drift_resume.py

@@ -0,0 +1,73 @@
+"""drift_resume.py — T16 §16P: drift-aware resume for months-later
+project re-entry.
+
+Months-later resume must NOT blindly trust the last checkpoint. The
+16O benchmark proved both cold and scaffold-assisted paths converge on
+the same principle: checkpoint claims NEVER substitute for reality.
+
+Flow:
+
+    load last verified scaffold state (expectations)
+        ↓
+    inspect current repo identity/state (observations)
+        ↓
+    compare expected vs actual  →  DriftRecord list
+        ↓
+    unchanged:  resume
+    drift:      classify EXPECTED / BENIGN / MATERIAL / CONFLICTING /
+                UNKNOWN — MATERIAL/UNKNOWN can block dependent work
+
+Reality wins unless the claim carries explicit user-locked authority
+(the one scaffold exception, already enforced by layout_contract).
+"""
+from __future__ import annotations
+
+from typing import Callable
+
+EXPECTED = "EXPECTED"      # checkpoint older than reality; benign aging
+BENIGN = "BENIGN"          # difference with no bearing on the claims
+MATERIAL = "MATERIAL"      # reality contradicts a scaffold claim
+CONFLICTING = "CONFLICTING"  # two authorities disagree; needs user
+UNKNOWN = "UNKNOWN"        # cannot decide — treated as blocking
+
+BLOCKING_CLASSES = (MATERIAL, CONFLICTING, UNKNOWN)
+
+
+def classify_drift(claim: dict, observation: dict) -> dict:
+    """Classify one claim-vs-observation pair. Deterministic:
+    verification evidence beats prose; explicit beats inferred.
+
+    claim:       {"claim": str, "kind": str, "verified_at": str,
+                  "user_locked": bool}
+    observation: {"actual": str|None, "checked_by": str,
+                  "matches": bool|None}
+    """
+    if observation.get("matches") is True:
+        return {"claim": claim.get("claim", ""),
+                "class": EXPECTED if claim.get("stale_ok") else "CONFIRMED",
+                "evidence": observation.get("checked_by", "")}
+    if observation.get("matches") is None:
+        return {"claim": claim.get("claim", ""), "class": UNKNOWN,
+                "evidence": f"unverifiable via {observation.get('checked_by', '?')}"}
+    if claim.get("user_locked"):
+        return {"claim": claim.get("claim", ""), "class": CONFLICTING,
+                "evidence": ("reality differs from a USER-LOCKED claim "
+                             "— user arbitration required")}
+    return {"claim": claim.get("claim", ""), "class": MATERIAL,
+            "evidence": (f"reality differs: {observation.get('actual', '')}"
+                         f" (checked by {observation.get('checked_by', '?')})")}
+
+
+def reconcile(claims: list[dict], observer: Callable[[dict], dict]) -> dict:
+    """Reconcile a set of checkpoint claims against reality.
+
+    observer(claim) → observation dict (runs the actual check: test
+    command, file read, git query — never trusts the claim text).
+
+    Returns {"resume": bool, "drifts": [...], "blocking": [...]}. The
+    resume decision: drift-aware, not blind — ANY blocking-class drift
+    stops dependent work until reconciled."""
+    drifts = [classify_drift(c, observer(c)) for c in claims]
+    blocking = [d for d in drifts if d["class"] in BLOCKING_CLASSES]
+    return {"resume": not blocking, "drifts": drifts,
+            "blocking": blocking}

src/rune/event_wait.py

@@ -0,0 +1,457 @@
+"""event_wait.py — T15 §15D: event-driven wait as the CANONICAL DEFAULT
+for long-running construction operations (builders, CI, deploys).
+
+GLOBAL WAITING INVARIANT (tranche brief):
+
+    EVENT DRIVES CONTINUATION.
+    TIMEOUT DETECTS FAILURE OF CONTINUATION.
+
+The W/R bridge (daemon.py + waitstore.py at RUNE_WR_ROOT, default
+~/.local/share/mimo-wait-resume) already provides: durable waits, event
+matching, per-session single-flight leases, process-tree ownership, and a
+resume driver. Until T15 its resume target was exclusively a MiMo
+session (global mimo-shaped resume_cmd), so Python controllers (stage
+drivers, builder sessions) could not be resumed by the bridge — which is
+exactly why fixed sleep/poll loops crept back into T12-T14.
+
+This adapter CLOSES that gap with the smallest possible surface:
+
+1. register_process_wait()  — a controller/operation registers a
+   PROCESS_EVENT wait (correlation: operation_id) whose per-wait
+   resume_cmd writes a resume token to a local FIFO. Registration is a
+   durable waitstore record: survives restarts, visible to reconciliation.
+
+2. run_builder_under_events() — launches the real builder command under
+   a notifier wrapper (separate session/process-group). The wrapper
+   observes the builder's actual exit and POSTs the completion event to
+   the daemon (/process-event). The wrapper is the event source; the
+   driver does NOT wait on the builder directly.
+
+3. await_resume_event() — the driver YIELDS: blocks in select() on the
+   FIFO (no polling loop, no sleep floor) with a watchdog deadline. The
+   daemon matches the event, advances the wait, and the per-wait
+   resume_cmd fires immediately — the driver resumes on the EVENT, not
+   after a configured delay.
+
+4. Watchdog semantics: on timeout the driver performs a ONE-SHOT
+   liveness inspection of the notifier process group (ALIVE → re-arm;
+   GONE → PROCESS_LOST recovery). The timeout is a failure detector,
+   never the scheduler.
+
+Idempotency (15E) is inherited from the bridge: match() applies only to
+WAITING_EXTERNAL waits, so duplicate/late events cannot double-resume;
+correlation by operation_id (run_ref) means a stale event from a prior
+operation matches nothing; the reconciler never verifies PROCESS_EVENT
+waits against Gitea (the event is the authority).
+"""
+
+from __future__ import annotations
+
+import importlib.util
+import json
+import os
+import select
+import signal
+import subprocess
+import sys
+import time
+from pathlib import Path
+from typing import Any
+
+DEFAULT_WR_ROOT = Path.home() / ".local" / "share" / "mimo-wait-resume"
+FALLBACK_WR_ROOT = Path("/home/nick/.local/share/mimo-wait-resume")
+DEFAULT_DAEMON_URL = "http://127.0.0.1:8791"
+
+# The notifier wrapper: runs the real builder command as its child,
+# observes the ACTUAL exit, and POSTs the completion event to the daemon.
+# It is the event SOURCE — deliberately independent of the waiting driver.
+_NOTIFIER_SRC = r"""
+import json, os, subprocess, sys, time, urllib.request
+op = sys.argv[1]; cmd = sys.argv[2:]
+start = time.time()
+proc = subprocess.run(cmd)
+completed_at = time.time()
+payload = {
+    "operation_id": op,
+    "exit_code": proc.returncode,
+    "completed_at": completed_at,
+    "started_at": start,
+}
+for attempt in range(3):
+    try:
+        req = urllib.request.Request(
+            os.environ["T15_DAEMON_URL"] + "/process-event",
+            data=json.dumps(payload).encode(),
+            headers={"Content-Type": "application/json"})
+        with urllib.request.urlopen(req, timeout=5) as r:
+            r.read()
+        break
+    except Exception as e:
+        if attempt == 2:
+            sys.stderr.write(f"event POST failed: {e}\n")
+            # best-effort: the watchdog is the safety net for lost events
+        time.sleep(1)
+"""
+
+
+def wr_root() -> Path:
+    env = os.environ.get("RUNE_WR_ROOT")
+    if env:
+        return Path(env)
+    if DEFAULT_WR_ROOT.exists():
+        return DEFAULT_WR_ROOT
+    return FALLBACK_WR_ROOT
+
+
+# waitstore instances cached per data root (isolated tests use per-test
+# tmp roots; the daemon uses its install dir). Loading via importlib.util
+# keeps the dynamic import pyright-clean — no unresolvable-import warning.
+_waitstore_cache: dict[str, Any] = {}
+
+
+def _load_waitstore() -> Any:
+    """Load the bridge's OWN waitstore module (never a copy), bound to
+    the adapter's data root. waitstore.py lives in the DAEMON CODE dir;
+    in production WR_ROOT is that dir (code + data together); in isolated
+    test environments WR_ROOT is a data-only dir, so the code dir is
+    resolved separately: RUNE_WR_CODE env → WR_ROOT → known installs."""
+    root = wr_root()
+    # waitstore binds its DATA dir from WR_ROOT at import time. Pin
+    # WR_ROOT to the adapter's authority so adapter-side and daemon-side
+    # state agree — otherwise a mismatch silently splits the wait store
+    # from the event matcher and continuation dies as a watchdog timeout.
+    os.environ["WR_ROOT"] = str(root)
+    cached = _waitstore_cache.get(str(root))
+    if cached is not None:
+        return cached
+    code_dir = None
+    for cand in (os.environ.get("RUNE_WR_CODE"), str(root),
+                 str(FALLBACK_WR_ROOT)):
+        if cand and (Path(cand) / "waitstore.py").is_file():
+            code_dir = Path(cand)
+            break
+    if code_dir is None:
+        raise RuntimeError(
+            "waitstore.py not found in known W/R roots "
+            "(RUNE_WR_CODE, WR_ROOT, ~/.local/share/mimo-wait-resume)")
+    spec = importlib.util.spec_from_file_location(
+        "waitstore", str(code_dir / "waitstore.py"))
+    if spec is None or spec.loader is None:
+        raise RuntimeError(f"cannot load waitstore from {code_dir}")
+    mod = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(mod)
+    _waitstore_cache[str(root)] = mod
+    return mod
+
+
+def register_process_wait(operation_id: str, *, contract_id: str = "",
+                          project: str = "", objective: str = "",
+                          watched_for: str = "",
+                          provider: str = "builder",
+                          event_type: str = "builder_exit",
+                          sha: str = "",
+                          watchdog_secs: int | None = None) -> dict:
+    """Register a durable PROCESS_EVENT wait for one operation. The
+    per-wait resume_cmd writes a resume token to the operation's FIFO —
+    the bridge can resume a Python controller, not just MiMo sessions.
+
+    provider/event_type are the adapter's event identity on the shared
+    bridge (16E: ONE wait primitive; builder / Gitea-CI / future adapters
+    differ ONLY in these identity fields + their liveness probe — the
+    registration, FIFO rendezvous, matching, idempotency, restart
+    recovery, and watchdog semantics are the same code for all).
+    Defaults preserve the T15 builder semantics."""
+    ws = _load_waitstore()
+    fifo = fifo_path(operation_id)
+    fifo.parent.mkdir(parents=True, exist_ok=True)
+    # Do NOT unlink+recreate an existing fifo: a caller that opened its
+    # read end before registering (open_fifo_reader, §16Y #10 race fix)
+    # would be orphaned onto the old inode while the daemon's resume
+    # writer opened the new one. The rendezvous is create-if-missing.
+    if not fifo.exists():
+        os.mkfifo(fifo)
+        os.chmod(fifo, 0o666)  # cross-user bridge (see open_fifo_reader)
+    writer = ("import sys,json,time,os;"
+              "fd=os.open(sys.argv[1], os.O_WRONLY | os.O_NONBLOCK);"
+              "os.write(fd, json.dumps({'wait_id': sys.argv[2],"
+              " 'resumed_at': time.time()}).encode());os.close(fd)")
+    rec = ws.new_wait(
+        session_id=f"controller-{operation_id}",
+        provider=provider,
+        event_type=event_type,
+        sha=sha,
+        repo=project or "construction",
+        run_ref=operation_id,
+        expected="success",
+        wait_kind="PROCESS_EVENT",
+        target_kind="headless",   # owned continuation target: FIFO writer
+        objective=objective or f"builder operation {operation_id}",
+        waited_for=watched_for or "builder process exit event",
+        verify_when_woken="reload canonical state; exit canary; "
+                          "independent verification",
+        on_success="continue programme continuation canary",
+        on_failure="classify failure; recovery or block",
+        resume_cmd=["python3", "-c", writer, str(fifo), "{wait_id}"],
+    )
+    rec["ext"] = {"contract_id": contract_id, "operation_id": operation_id,
+                  "fifo": str(fifo),
+                  "watchdog_secs": watchdog_secs
+                  or int(os.environ.get("WR_WATCH_SECS", "600"))}
+    ws.save(rec)
+    # cross-user bridge (T16 §16C live finding): the daemon reconciler may
+    # run under a different account than the registering controller — the
+    # durable record must be readable/writable on both sides
+    try:
+        os.chmod(ws.WAITS / f"{rec['wait_id']}.json", 0o666)
+    except OSError:
+        pass  # single-user deployment: nothing to relax
+    return rec
+
+
+def fifo_path(operation_id: str) -> Path:
+    return wr_root() / "resumes" / f"{operation_id}.fifo"
+
+
+def open_fifo_reader(operation_id: str) -> int:
+    """Open the operation's FIFO read end (O_NONBLOCK) BEFORE the wait is
+    registered. §16Y adversarial #10: an event that arrives after
+    registration but before the controller enters select() would find no
+    reader — the resume write fails (ENXIO) and the wait lands
+    RESUME_UNVERIFIED while the controller sleeps until the watchdog.
+    Holding the reader open across the registration window closes that
+    race for every adapter; the fd is passed to await_resume_event and
+    closed by the caller."""
+    fifo = fifo_path(operation_id)
+    fifo.parent.mkdir(parents=True, exist_ok=True)
+    if fifo.exists():
+        fifo.unlink()
+    os.mkfifo(fifo)
+    # CROSS-USER BRIDGE (T17 17C root-cause, re-landed): the T16 chmod was
+    # lost in a rebase resolution — FIFOs landed 644 (666 & umask 022),
+    # and the daemon running as a DIFFERENT user got EACCES on the resume
+    # write (rc=1 → RESUME_UNVERIFIED) while same-user tests passed.
+    # chmod AFTER mkfifo is the only reliable act — umask never applies.
+    os.chmod(fifo, 0o666)
+    return os.open(fifo, os.O_RDONLY | os.O_NONBLOCK)
+
+
+def run_builder_under_events(builder_cmd: list[str], operation_id: str, *,
+                             daemon_url: str = DEFAULT_DAEMON_URL):
+    """Launch the REAL builder under the event notifier (separate process
+    group). Returns the Popen for the notifier (pgid == its pid). The
+    driver must NOT wait on this directly — it yields via
+    await_resume_event; the notifier's exit event comes back through the
+    bridge."""
+    env = dict(os.environ, T15_DAEMON_URL=daemon_url)
+    log_file = wr_root() / "resumes" / f"{operation_id}.builder.log"
+    log_file.parent.mkdir(parents=True, exist_ok=True)
+    with open(log_file, "w") as lf:
+        return subprocess.Popen(
+            [sys.executable, "-c", _NOTIFIER_SRC, operation_id, *builder_cmd],
+            stdout=lf, stderr=subprocess.PIPE,
+            text=True, start_new_session=True, env=env)
+
+
+def _pgroup_alive(pgid: int) -> bool:
+    """Liveness = at least one NON-ZOMBIE member of the process group.
+    A killed-but-unreaped notifier is a zombie and does NOT count —
+    treating zombies as alive would misclassify PROCESS_LOST as alive."""
+    try:
+        os.killpg(pgid, 0)  # group exists at all?
+    except (ProcessLookupError, PermissionError):
+        return False
+    for proc in Path("/proc").iterdir():
+        if not proc.name.isdigit():
+            continue
+        try:
+            stat = (proc / "stat").read_text()
+            # field 5 is state; comm (field 2) may contain spaces/parens
+            state = stat.rsplit(")", 1)[1].split()[0]
+            pgid_field = int(stat.rsplit(")", 1)[1].split()[2])
+        except (OSError, ValueError, IndexError):
+            continue
+        if pgid_field == pgid and state != "Z":
+            return True
+    return False
+
+
+def _cmdline_alive(needle: str) -> bool:
+    """Liveness fallback: any live process whose cmdline contains the
+    needle. Real builders (node CLIs) may re-group their children, so
+    pgid alone can false-negative — T15Q observed a 65-minute real build
+    outliving the watchdog AND invisible to the pgid scan. The needle
+    should be operation-specific (the prompt marker), not generic."""
+    for proc in Path("/proc").iterdir():
+        if not proc.name.isdigit():
+            continue
+        try:
+            cmdline = (proc / "cmdline").read_bytes().decode(
+                errors="replace")
+            stat = (proc / "stat").read_text()
+            state = stat.rsplit(")", 1)[1].split()[0]
+        except (OSError, ValueError, IndexError):
+            continue
+        if needle in cmdline and state != "Z":
+            return True
+    return False
+
+
+def liveness(operation_id: str, pgid: int | None = None) -> bool:
+    """Watchdog helper: is the watched operation actually alive? pgid
+    membership first; cmdline evidence as the regroup-tolerant fallback."""
+    if pgid and _pgroup_alive(pgid):
+        return True
+    return _cmdline_alive(f"op:{operation_id}")
+
+
+def await_resume_event(operation_id: str, *, watchdog_secs: int,
+                       pgid: int | None = None,
+                       max_rearms: int = 20,
+                       fifo_fd: int | None = None) -> dict:
+    """YIELD until the bridge resumes this controller. Blocks in select()
+    on the FIFO — no sleep, no polling.
+
+    fifo_fd: a pre-opened read end from open_fifo_reader() (§16Y #10 race
+    fix — the fd must already exist when the wait becomes registered).
+    Caller-owned fds are never closed here; on EOF (writer gone, no
+    token) the call returns {"kind": "eof"} instead of busy-looping on an
+    always-readable fd — the caller falls back to the durable waitstore
+    record via resume_or_recover/operation_wait.
+
+    Watchdog semantics (§15D): the deadline is a FAILURE DETECTOR, never
+    the scheduler. On expiry, liveness is inspected ONCE (pgid membership,
+    then operation-specific cmdline evidence — real builders regroup their
+    children): ALIVE → re-arm (select again, up to max_rearms); GONE →
+    PROCESS_LOST. The T15Q live run proved why: a real 65-minute model
+    build outlived a single 900s deadline, and node regrouping hid the
+    processes from the pgid scan — neither is a failure of continuation.
+    """
+    owned_fd = fifo_fd is None
+    if owned_fd:
+        fifo = fifo_path(operation_id)
+        # idempotent re-arm: a prior await consumed+unlinked the fifo; the
+        # rendezvous point is (re)created so watchdog re-arm and recovery
+        # paths can wait again on the same operation
+        fifo.parent.mkdir(parents=True, exist_ok=True)
+        if not fifo.exists():
+            os.mkfifo(fifo)
+        fd = os.open(fifo, os.O_RDONLY | os.O_NONBLOCK)
+    else:
+        fd = fifo_fd
+    try:
+        for arm in range(max_rearms + 1):
+            deadline = time.monotonic() + watchdog_secs
+            while True:
+                remaining = deadline - time.monotonic()
+                if remaining <= 0:
+                    break
+                r, _, _ = select.select([fd], [], [], remaining)
+                if not r:
+                    continue
+                data = os.read(fd, 4096)
+                if not data:
+                    if owned_fd:
+                        continue
+                    # caller-owned fd at EOF: writer is gone without
+                    # delivering a token — never a busy loop; the durable
+                    # record decides what actually happened
+                    return {"kind": "eof"}
+                try:
+                    token = json.loads(data.decode())
+                except ValueError:
+                    break  # malformed token → treat as expiry
+                ws = _load_waitstore()
+                rec = ws.load(token.get("wait_id", ""))
+                return {"kind": "event", "wait": rec,
+                        "resumed_at": token.get("resumed_at")}
+            # deadline expired: liveness decides re-arm vs PROCESS_LOST
+            alive = liveness(operation_id, pgid)
+            if not alive:
+                return {"kind": "watchdog", "alive": False,
+                        "rearms": arm}
+        return {"kind": "watchdog", "alive": liveness(operation_id, pgid),
+                "rearms": max_rearms}
+    finally:
+        if owned_fd:
+            os.close(fd)
+
+
+def operation_wait(operation_id: str) -> dict | None:
+    ws = _load_waitstore()
+    for rec in ws.all_waits():
+        if rec.get("ext", {}).get("operation_id") == operation_id:
+            return rec
+    return None
+
+
+def resume_or_recover(operation_id: str) -> dict:
+    """15E case 10 — controller restart after the event was persisted but
+    before the controller handled it. The durable waitstore record is the
+    authority, not the transient FIFO token: if the wait already reached a
+    terminal/resumable state, return it directly (idempotent recovery, no
+    duplicate transition); if still WAITING_EXTERNAL, re-arm the normal
+    event wait."""
+    rec = operation_wait(operation_id)
+    if rec is None:
+        return {"kind": "unknown_operation"}
+    if rec.get("status") in ("COMPLETED", "FAILED", "READY_TO_RESUME",
+                             "READY_TO_RETRY", "RESUMING"):
+        # RESUMING counts as recovered: the resume delivery is in flight
+        # (or already consumed the rendezvous). The durable record is the
+        # truth; the controller restarts from it, never from the token.
+        return {"kind": "event", "wait": rec, "recovered": True}
+    if rec.get("status") == "RESUME_UNVERIFIED" and \
+            rec.get("process_event_evidence"):
+        # the completion EVENT was processed (verdict recorded durably);
+        # only the resume-token delivery was unverified (e.g. controller
+        # was down, so no reader consumed the rendezvous). Canonical
+        # state wins — this IS the restart-after-event case.
+        return {"kind": "event", "wait": rec, "recovered": True}
+    return {"kind": "still_waiting", "wait": rec}
+
+
+def find_matching_event(provider: str, event_type: str, run_ref: str) -> dict | None:
+    """Search the bridge's durable event log for an event matching this
+    identity. T16 §16C live finding: on a serialized single-runner
+    machine the deploy job ALWAYS starts after ci.yml completes — the
+    completion event lands before the waiter registers, and a pure
+    forward-only FIFO rendezvous can never deliver it (both production
+    cycles fell back to the 900s deadline). Events are durable precisely
+    so they can be replayed: a waiter that finds a matching event at
+    registration goes straight to the authority read instead of sleeping
+    for a wake that already happened."""
+    import json as _json
+    ws = _load_waitstore()
+    events_dir = ws.EVENTS
+    if not events_dir.is_dir():
+        return None
+    best = None
+    for f in sorted(events_dir.glob("*.json"),
+                    key=lambda p: p.stat().st_mtime, reverse=True):
+        try:
+            rec = _json.loads(f.read_text())
+        except (OSError, ValueError):
+            continue
+        ev = rec.get("event", {})
+        if (ev.get("provider") == provider
+                and ev.get("event_type") == event_type
+                and str(ev.get("run_ref", "")) == run_ref):
+            best = ev  # newest match wins
+            break
+    return best
+
+
+def post_process_event(operation_id: str, exit_code: int, *,
+                       daemon_url: str = DEFAULT_DAEMON_URL) -> dict:
+    """Direct event injection (test seam + manual reconciliation). The
+    normal path is the notifier wrapper; this exists so tests and
+    watchdog recovery can emit the same event shape."""
+    import urllib.request
+    payload = {"operation_id": operation_id, "exit_code": exit_code,
+               "completed_at": time.time()}
+    req = urllib.request.Request(
+        daemon_url + "/process-event",
+        data=json.dumps(payload).encode(),
+        headers={"Content-Type": "application/json"})
+    with urllib.request.urlopen(req, timeout=5) as r:
+        return json.loads(r.read())

src/rune/evidence_probe.py

@@ -192,7 +192,8 @@ def _default_runner(prompt: str, artifact_path: str) -> None:
     import subprocess
 
     mimo_bin = os.environ.get("MIMO_BIN", "/var/lib/gitea-desktop-runner/bin/mimo")
-    mimo_model = os.environ.get("MIMO_MODEL", "Verboo Code/glm-5.3-flash")
+    from rune.provider_context import require_model
+    mimo_model = require_model()  # T16 §16F: provider layer owns selection
     cmd = [
         mimo_bin,
         "run",

src/rune/failure_recovery.py

@@ -0,0 +1,108 @@
+"""failure_recovery.py — T15 §15T: failure classification + recovery
+policy.
+
+Before retrying, classify the failure:
+
+    TRANSIENT            bounded immediate retry appropriate
+    INTERMITTENT         bounded retry/backoff
+    KNOWN_PROCESS_DEFECT apply evidenced repair, then retry
+    PERMANENT / INVALID  do NOT retry blindly
+    APPROVAL_REQUIRED    block
+    UNKNOWN              bounded evidence probe, then block/escalate
+
+No infinite retry loops: every class carries a retry budget, and
+PERMANENT/APPROVAL_REQUIRED never auto-retry.
+"""
+
+from __future__ import annotations
+
+TRANSIENT = "TRANSIENT"
+INTERMITTENT = "INTERMITTENT"
+KNOWN_PROCESS_DEFECT = "KNOWN_PROCESS_DEFECT"
+PERMANENT = "PERMANENT"
+APPROVAL_REQUIRED = "APPROVAL_REQUIRED"
+PROVIDER_UNAVAILABLE = "PROVIDER_UNAVAILABLE"
+UNKNOWN = "UNKNOWN"
+
+CLASSES = (TRANSIENT, INTERMITTENT, KNOWN_PROCESS_DEFECT, PERMANENT,
+           APPROVAL_REQUIRED, PROVIDER_UNAVAILABLE, UNKNOWN)
+
+# recovery policy per class: (max_retries, action)
+POLICY = {
+    TRANSIENT: (2, "retry_immediately"),
+    INTERMITTENT: (3, "retry_with_backoff"),
+    KNOWN_PROCESS_DEFECT: (1, "apply_evidenced_repair_then_retry"),
+    PERMANENT: (0, "do_not_retry"),
+    APPROVAL_REQUIRED: (0, "block"),
+    # T16 §16F: provider failures NEVER auto-retry and NEVER fall back
+    # to another provider — preserve the real error, block the branch.
+    PROVIDER_UNAVAILABLE: (0, "block_branch_preserve_error"),
+    UNKNOWN: (1, "bounded_evidence_probe_then_block_or_escalate"),
+}
+
+# deterministic markers (substring, lowercase) — the classifier is
+# evidence-first: explicit signals win; absence defaults to UNKNOWN
+_TRANSIENT_MARKERS = (
+    "connection reset", "timeout", "timed out", "temporarily unavailable",
+    "rate limit", "429", "503", "econnreset", "socket hang up",
+    "service unavailable",
+)
+_INTERMITTENT_MARKERS = (
+    "flaky", "intermittent", "occasionally fails", "sometimes fails",
+    "retry succeeded",
+)
+_PERMANENT_MARKERS = (
+    "invalid input", "schema-invalid", "not found", "no such file",
+    "permission denied", "unsupported", "incompatible",
+    "module not found", "syntaxerror", "typeerror",
+)
+_APPROVAL_MARKERS = (
+    "approval required", "needs authorization", "user confirmation",
+    "destructive", "external service", "new repository",
+    "user-locked", "user locked",
+)
+_KNOWN_DEFECT_MARKERS = (
+    "stale document", "forgotten save", "wrong path", "guessed api",
+    "missing import", "no-op edit", "substring match", "wrong signature",
+)
+
+
+def classify_failure(error_text: str, *, context: dict | None = None) -> dict:
+    """Classify one failure into a recovery class. Deterministic: marker
+    evidence first; explicit context hints (e.g. known defect id) second;
+    UNKNOWN when nothing matches (never manufactured certainty)."""
+    text = (error_text or "").lower()
+    context = context or {}
+    known_defect = context.get("known_defect_id")
+    if known_defect:
+        return {"class": KNOWN_PROCESS_DEFECT,
+                "evidence": f"known defect: {known_defect}",
+                "policy": POLICY[KNOWN_PROCESS_DEFECT]}
+    # T16 §16F: an explicit provider-context failure wins over every
+    # generic marker — the real status/API code is preserved verbatim in
+    # the evidence, the branch blocks, and no fallback is attempted.
+    if context.get("provider_error"):
+        return {"class": PROVIDER_UNAVAILABLE,
+                "evidence": (f"provider error preserved: {error_text}"
+                             if error_text else "provider error"),
+                "policy": POLICY[PROVIDER_UNAVAILABLE]}
+    if any(m in text for m in _APPROVAL_MARKERS):
+        return {"class": APPROVAL_REQUIRED,
+                "evidence": "approval marker present",
+                "policy": POLICY[APPROVAL_REQUIRED]}
+    if any(m in text for m in _PERMANENT_MARKERS):
+        return {"class": PERMANENT, "evidence": "permanent marker present",
+                "policy": POLICY[PERMANENT]}
+    if any(m in text for m in _KNOWN_DEFECT_MARKERS):
+        return {"class": KNOWN_PROCESS_DEFECT,
+                "evidence": "known process-defect marker present",
+                "policy": POLICY[KNOWN_PROCESS_DEFECT]}
+    if any(m in text for m in _INTERMITTENT_MARKERS):
+        return {"class": INTERMITTENT,
+                "evidence": "intermittent marker present",
+                "policy": POLICY[INTERMITTENT]}
+    if any(m in text for m in _TRANSIENT_MARKERS):
+        return {"class": TRANSIENT, "evidence": "transient marker present",
+                "policy": POLICY[TRANSIENT]}
+    return {"class": UNKNOWN, "evidence": "no marker matched",
+            "policy": POLICY[UNKNOWN]}

src/rune/programme_continuation.py

@@ -0,0 +1,349 @@
+"""programme_continuation.py — T15 §15M: the PROGRAMME CONTINUATION
+CANARY.
+
+The user observation this answers: subtask completion repeatedly ended
+whole programmes. "task done" is not "programme done".
+
+After ANY stage/subtask reports complete, the runner asks one bounded
+question — is the overall construction/programme objective satisfied?
+If not, and eligible authorized stages remain, the NEXT stage starts
+immediately (no user re-prompt, no voluntary stop). Blocked branches
+fall back to pre-authorized independent work.
+
+Boundary (tranche brief): this is a bounded CONSTRUCTION-programme rule
+over Scaffolding stage contracts — NOT the future general-purpose
+Substrate scheduler. It knows nothing outside one programme's stage
+list.
+"""
+
+from __future__ import annotations
+
+import json
+import time
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Any, cast
+
+
+# stage states
+PENDING = "PENDING"
+ELIGIBLE = "ELIGIBLE"
+WAITING_EVENT = "WAITING_EVENT"
+COMPLETE = "COMPLETE"
+BLOCKED = "BLOCKED"
+APPROVAL_REQUIRED = "APPROVAL_REQUIRED"
+FAILED = "FAILED"
+
+TERMINAL = {COMPLETE, BLOCKED, APPROVAL_REQUIRED, FAILED}
+
+
+@dataclass
+class ProgrammeStage:
+    stage_id: str
+    objective: str
+    state: str = PENDING
+    depends_on: list = field(default_factory=list)   # stage_ids
+    priority: int = 0                                # lower = sooner
+    result: dict = field(default_factory=dict)
+
+    def to_dict(self) -> dict:
+        return {"stage_id": self.stage_id, "objective": self.objective,
+                "state": self.state, "depends_on": list(self.depends_on),
+                "priority": self.priority}
+
+
+@dataclass
+class Programme:
+    """One bounded construction programme: its objective, its authorized
+    stages, and the continuation rule. Persisted to the contract store's
+    project area (durable, reloadable, no transcript dependency)."""
+    programme_id: str
+    objective: str
+    stages: list = field(default_factory=list)  # [ProgrammeStage]
+    fallbacks: list = field(default_factory=list)  # pre-authorized ids
+
+    # ── the continuation canary (15M) ────────────────────────────────
+    def continuation_canary(self) -> dict:
+        """'task done' ≠ 'programme done'. Returns the verdict AND the
+        next action — the runner never has to guess whether to stop."""
+        complete = [s for s in self.stages if s.state == COMPLETE]
+        if len(complete) == len(self.stages):
+            return {"verdict": "PROGRAMME_COMPLETE",
+                    "next_action": None,
+                    "reason": f"all {len(self.stages)} stages complete"}
+        eligible = self.eligible_stages()
+        if eligible:
+            nxt = eligible[0]
+            return {"verdict": "PROGRAMME_OPEN",
+                    "next_action": ("start", nxt.stage_id),
+                    "reason": (f"{len(complete)}/{len(self.stages)} "
+                               f"complete; {len(eligible)} eligible — "
+                               f"continue immediately")}
+        waiting = [s for s in self.stages
+                   if s.state == WAITING_EVENT]
+        if waiting:
+            return {"verdict": "PROGRAMME_OPEN_WAITING",
+                    "next_action": ("await_event", waiting[0].stage_id),
+                    "reason": "event-driven wait active — no polling"}
+        blocked = [s for s in self.stages if s.state == BLOCKED]
+        if blocked and self.fallbacks:
+            return {"verdict": "PROGRAMME_OPEN",
+                    "next_action": ("fallback", self.fallbacks[0]),
+                    "reason": (f"{len(blocked)} blocked; pre-authorized "
+                               f"fallback available")}
+        if blocked:
+            return {"verdict": "ALL_REMAINING_BLOCKED",
+                    "next_action": None,
+                    "reason": f"{len(blocked)} stage(s) blocked, no fallback"}
+        approval = [s for s in self.stages
+                    if s.state == APPROVAL_REQUIRED]
+        if approval:
+            return {"verdict": "APPROVAL_REQUIRED",
+                    "next_action": None,
+                    "reason": (f"{len(approval)} stage(s) need user "
+                               f"authorization")}
+        # T17 §17E: all stages terminal but some FAILED — the programme
+        # may still close (programme-level acceptance decides), never
+        # hang in UNKNOWN_INTERRUPTION.
+        if all(s.state in TERMINAL for s in self.stages):
+            failed = [s.stage_id for s in self.stages if s.state == FAILED]
+            return {"verdict": "PROGRAMME_COMPLETE_WITH_FAILURES",
+                    "next_action": None,
+                    "reason": (f"all stages terminal; failed: {failed} — "
+                               f"programme-level acceptance decides")}
+        return {"verdict": "UNKNOWN_INTERRUPTION", "next_action": None,
+                "reason": "no stage runnable and none terminal"}
+
+    def eligible_stages(self) -> list:
+        """PENDING stages whose dependencies are COMPLETE, priority-first.
+        ELIGIBLE counts as runnable too — an eligible stage nobody started
+        is exactly the case the continuation canary exists to catch."""
+        done = {s.stage_id for s in self.stages if s.state == COMPLETE}
+        out = [s for s in self.stages
+               if s.state in (PENDING, ELIGIBLE)
+               and all(d in done for d in s.depends_on)]
+        out.sort(key=lambda s: (s.priority, s.stage_id))
+        return out
+
+    # ── durable persistence (canonical store, not transcript) ────────
+    def save(self, base: Path) -> Path:
+        base.mkdir(parents=True, exist_ok=True)
+        p = base / f"{self.programme_id}.json"
+        p.write_text(json.dumps({
+            "programme_id": self.programme_id,
+            "objective": self.objective,
+            "stages": [s.to_dict() for s in self.stages],
+            "fallbacks": list(self.fallbacks),
+            "saved_at": time.strftime("%Y-%m-%dT%H:%M:%SZ",
+                                      time.gmtime()),
+        }, indent=1))
+        return p
+
+    @classmethod
+    def load(cls, base: Path, programme_id: str) -> "Programme | None":
+        p = base / f"{programme_id}.json"
+        if not p.exists():
+            return None
+        d = json.loads(p.read_text())
+        prog = cls(programme_id=d["programme_id"],
+                   objective=d["objective"], fallbacks=d.get("fallbacks", []))
+        for s in d["stages"]:
+            prog.stages.append(ProgrammeStage(**s))
+        return prog
+
+
+def run_programme(prog: Programme, stage_runner, *,
+                  max_transitions: int = 100) -> list[dict]:
+    """Drive one programme to a legitimate stop. `stage_runner(stage)` —
+    provided by the caller — must return a state for the stage. Between
+    EVERY stage: persist → continuation canary → next. Returns the
+    transition log (the 15M proof artifact)."""
+    log: list[dict] = []
+    for _ in range(max_transitions):
+        prog_canary = prog.continuation_canary()
+        log.append({"canary": prog_canary})
+        verdict, action = prog_canary["verdict"], prog_canary["next_action"]
+        if action is None:
+            break  # legitimate stop — reason recorded in the log
+        kind, target = action
+        if kind == "fallback":
+            log.append({"fallback_started": target})
+            break
+        stage = next(s for s in prog.stages if s.stage_id == target)
+        stage.state = stage_runner(stage)   # runner decides the new state
+        log.append({"stage": target, "new_state": stage.state})
+    return log
+
+
+# ── T17 §17D: production wiring — Scaffolding-integrated driver ─────────
+#
+# The 15M canary proved the RULE; this adapter makes it CANONICAL. One
+# programme = a list of stages; each stage maps 1:1 to a Scaffolding
+# contract (stage_id == contract_id). Stages run through the Construction
+# Interface lifecycle (approved→building→verifying→closed); asynchronous
+# stage work runs under the event bridge (WAITING_EVENT is a state, not
+# a stop). Persisted to the scaffold store's programme area — reloadable
+# after restart with no transcript.
+
+DEFAULT_PROGRAMME_AREA = "programmes"
+
+
+def _programme_base(scaffold_store: Path) -> Path:
+    return Path(scaffold_store) / DEFAULT_PROGRAMME_AREA
+
+
+def scaffolding_stage_runner(interface, *, scaffold_store: Path,
+                             daemon_url: str = "",
+                             watchdog_secs: int = 900):
+    """Return a stage_runner for run_programme bound to the Construction
+    Interface. The stage's work is read from stage.result["work"] — a
+    callable (interface, stage) → dict with {"ok": bool, "evidence": str,
+    "async_cmd": [argv] optional}. When async_cmd is present the work
+    runs under the event bridge and the driver resumes on the EVENT;
+    canonical verification still decides completion truth. Stages without
+    work complete as governance-only transitions."""
+    from rune import event_wait as ew  # local import: optional dependency
+
+    def _phase(contract_id: str) -> str:
+        # ALWAYS re-read: transition() saves through its own doc copy, so
+        # a locally-held contract dict goes stale after any transition
+        # (T17 17E live finding — the draft→approved→building chain
+        # silently stopped after the first step on a stale reference).
+        doc = interface.scaffold.load()
+        c = interface.scaffold.contract(doc, contract_id)
+        return (c or {}).get("phase", "")
+
+    def run(stage: ProgrammeStage) -> str:
+        contract_id = stage.stage_id
+        phase = _phase(contract_id)
+        if phase == "":
+            stage.result["error"] = f"no contract {contract_id}"
+            return FAILED
+        if phase == "closed":
+            return COMPLETE  # idempotent re-entry (duplicate signal case)
+        if phase == "draft":
+            interface.transition(contract_id, "approved",
+                                 evidence_ref="programme-stage-start")
+        if _phase(contract_id) == "approved":
+            interface.transition(contract_id, "building",
+                                 evidence_ref="programme-stage-start")
+        work = cast("Any", stage.result.get("work"))
+        if work is None:
+            interface.transition(contract_id, "verifying",
+                                 evidence_ref="governance-only-stage")
+            interface.close(
+                contract_id,
+                observed_evidence=[{"source": "programme-driver",
+                                    "detail": "governance-only stage: "
+                                              "no work, closure by "
+                                              "programme acceptance"}],
+                scope_files=[])
+            return COMPLETE
+        # work is either a callable (interface, stage) → outcome dict, or
+        # a literal outcome spec dict ({"ok", "evidence", "async_cmd"?})
+        outcome: dict = (dict(cast("dict", work(interface, stage)))
+                         if callable(work) else dict(cast("dict", work)))
+        async_cmd = outcome.get("async_cmd")
+        if async_cmd:
+            op = f"stage-{contract_id}-{int(time.time())}"
+            ew.register_process_wait(
+                op, contract_id=contract_id,
+                project=getattr(interface.scaffold, "project_id", ""),
+                objective=stage.objective,
+                watched_for="stage work completion event",
+                watchdog_secs=watchdog_secs)
+            notifier = ew.run_builder_under_events(
+                async_cmd, op, daemon_url=daemon_url or ew.DEFAULT_DAEMON_URL)
+            stage.result["operation_id"] = op
+            stage.result["notifier_pgid"] = notifier.pid
+            result = ew.await_resume_event(op, watchdog_secs=watchdog_secs,
+                                           pgid=notifier.pid)
+            if result["kind"] != "event":
+                stage.result["error"] = f"stage event wait: {result['kind']}"
+                interface.transition(contract_id, "failed",
+                                     evidence_ref="stage-event-lost")
+                return FAILED
+            rec: dict = result.get("wait") or {}
+            outcome["evidence"] = (f"event resume; wait {rec.get('wait_id')} "
+                                   f"verdict={rec.get('verdict')}")
+        if not outcome.get("ok"):
+            stage.result["error"] = outcome.get("evidence", "work failed")
+            # truthful failure: record the evidence ON the contract and
+            # drive it to the failed phase via the legal transition map
+            # (building→failed is legal; no exception swallowing)
+            doc = interface.scaffold.load()
+            c = interface.scaffold.contract(doc, contract_id)
+            if c is not None:
+                c.setdefault("checkpoints", []).append({
+                    "at": time.strftime("%Y-%m-%dT%H:%M:%SZ",
+                                        time.gmtime()),
+                    "phase": c.get("phase", ""),
+                    "evidence": "programme-stage-failed",
+                    "note": (outcome.get("evidence") or "work failed")[:200],
+                })
+                interface.scaffold.save(doc)
+                phase = c.get("phase", "")
+                if phase == "building":
+                    interface.transition(contract_id, "failed",
+                                         evidence_ref="programme-stage-failed")
+            return FAILED
+        # completion truth: work evidence recorded ON the contract, then
+        # closed (evidence lives on the durable object, not the log)
+        if _phase(contract_id) == "building":
+            interface.transition(contract_id, "verifying",
+                                 evidence_ref="programme-stage-verify")
+        doc = interface.scaffold.load()
+        c = interface.scaffold.contract(doc, contract_id)
+        if c is not None:
+            c.setdefault("checkpoints", []).append({
+                "at": time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
+                "phase": "verifying",
+                "evidence": "programme-stage-work",
+                "note": (outcome.get("evidence") or "work completed")[:200],
+            })
+            interface.scaffold.save(doc)
+        interface.close(
+            contract_id,
+            observed_evidence=[{"source": "programme-driver",
+                                "detail": outcome.get("evidence", "ok")}],
+            scope_files=[])
+        return COMPLETE
+
+    return run
+
+
+def run_construction_programme(prog: Programme, interface, *,
+                               scaffold_store: Path,
+                               daemon_url: str = "",
+                               watchdog_secs: int = 900,
+                               max_transitions: int = 100) -> list[dict]:
+    """Canonical multi-stage construction continuation (T17 §17D):
+    run_programme with the Scaffolding-bound runner; persists the
+    programme around every transition so a restart resumes from the
+    durable record, never a transcript. Returns the transition log."""
+    runner = scaffolding_stage_runner(
+        interface, scaffold_store=scaffold_store,
+        daemon_url=daemon_url, watchdog_secs=watchdog_secs)
+    log: list[dict] = []
+    base = _programme_base(scaffold_store)
+    for _ in range(max_transitions):
+        prog.save(base)
+        canary = prog.continuation_canary()
+        log.append({"canary": canary})
+        verdict, action = canary["verdict"], canary["next_action"]
+        if action is None:
+            break
+        kind, target = action
+        if kind == "fallback":
+            log.append({"fallback_started": target})
+            break
+        stage = next(s for s in prog.stages if s.stage_id == target)
+        if stage.state == COMPLETE:
+            continue  # duplicate continuation signal: idempotent
+        stage.state = ELIGIBLE
+        new_state = runner(stage)
+        stage.state = new_state
+        log.append({"stage": target, "new_state": new_state})
+        prog.save(base)
+    prog.save(base)
+    return log

src/rune/provider_context.py

@@ -0,0 +1,48 @@
+"""provider_context.py — T16 §16F: provider ownership boundary.
+
+ARCHITECTURE RULE (T16 mandate, permanent):
+
+    SCAFFOLDING DOES NOT OWN PROVIDER SELECTION.
+
+Provider, model, provider fallback, provider source, and provider account
+belong to the surrounding Rune/TUI/provider layer. Construction code never
+names a provider. The ONLY supported way for a driver/probe (builder
+session, research stage, evidence probe — all PROVIDER_LAYER callers) to
+learn which model to use is the ambient provider context, i.e. the
+environment the surrounding layer sets.
+
+    works → continue
+    missing/unresolvable → fail LOUDLY with PROVIDER_UNRESOLVED
+    provider call fails → preserve the REAL error and status/API code,
+        classify BLOCKED/PROVIDER_UNAVAILABLE, stop the dependent branch
+
+NO silent default provider. NO automatic fallback. NO preference mutation.
+(Automatic provider fallback is DEFERRED to the future multi-source /
+local-model architecture.)
+"""
+from __future__ import annotations
+
+import os
+
+
+class ProviderUnresolved(RuntimeError):
+    """Raised when no provider context is available and a silent default
+    would otherwise have been selected. Deliberately loud: provider
+    selection is a user-layer decision, never a code default."""
+
+
+def require_model(env_var: str = "MIMO_MODEL") -> str:
+    """Resolve the current model from the ambient provider context.
+
+    The surrounding provider layer (TUI / operator / session environment)
+    owns this variable. A driver or probe MUST NOT carry its own default:
+    a missing value is a configuration gap the user layer must fix, not
+    something construction code may paper over by picking a provider."""
+    model = (os.environ.get(env_var) or "").strip()
+    if not model:
+        raise ProviderUnresolved(
+            f"PROVIDER_UNRESOLVED: {env_var} is not set — provider/model "
+            f"selection belongs to the provider layer (set {env_var} in "
+            f"the calling environment); no silent default is permitted "
+            f"(T16 §16F)")
+    return model

src/rune/rune_operator.py

@@ -787,6 +787,20 @@ class RuneOperator:
         NO duplicate intelligence here."""
         from .construction_mode import classify_mode, _STANDALONE_WORDS
         from .construction_activation import activate
+        # §15J case F: manual override feeds the SAME decision path — a
+        # leading "/scaffold-mode <MODE>" (or "scaffold-mode: <MODE>")
+        # marker is stripped and passed as explicit_override with
+        # override_authority="user". NOT a second implementation:
+        # classification still runs through classify_mode with provenance.
+        override = None
+        m = re.match(
+            r"^(?:/scaffold-mode|scaffold-mode:)\s+"
+            r"(NOT_CONSTRUCTION|LOCAL_CHANGE|EXISTING_PROJECT_CHANGE|"
+            r"PROJECT_BUILD|UNKNOWN)\s+(.*)$",
+            description, re.DOTALL)
+        if m:
+            override = m.group(1)
+            description = m.group(2).strip()
         low = description.lower()
         tokens = set(__import__("re").split(r"[^a-z0-9]+", low))
         # §12D: the NATIVE hook must not force the Rune target — standalone
@@ -802,7 +816,9 @@ class RuneOperator:
             target, exists = None, None    # unknown → probes, not invention
         mode_decision = classify_mode(
             description, target_project=target, project_exists=exists,
-            affected_surface=[])
+            affected_surface=[],
+            explicit_override=override,
+            override_authority="user" if override else None)
         from .scaffold import Scaffold
         s = Scaffold(self._construction_store(), "issue161-harness")
         doc = s.load()

src/rune/rune_tui.py

@@ -8,7 +8,7 @@ mode system, subject header, shared memory and identity.
 
 This is not an interface. This is how I see, speak, and interact with the world.
 
-Version: v0.66.58
+Version: v0.66.68
 Features: 41 (24 original + 17 new native commands)
 
 Usage:
@@ -1856,7 +1856,7 @@ class RuneApp(App):
         # Show version info
         self.chat_channel.add_message(
             "system",
-            f"[dim]Rune v0.66.58 | Textual TUI | {len(loaded)} native modules[/dim]",
+            f"[dim]Rune v0.66.68 | Textual TUI | {len(loaded)} native modules[/dim]",
         )
 
         # Auto-show night mode suggestion if evening/night hours

src/rune/stage_guards.py

@@ -0,0 +1,281 @@
+"""stage_guards.py — T15 §15F–§15I: guarded stage execution for Scaffolding.
+
+Design lessons absorbed (NOT the systems they came from): pre/post-
+conditions and invariants (Design by Contract), readiness gating of
+downstream work (DAGs), distinct startup/liveness/exit probe meanings
+(K8s probes), desired-vs-actual reconciliation (K8s controllers),
+idempotent retry + failure classification (Temporal), explicit
+inputs/outputs and null-build checks (Bazel), policy decision separated
+from the governed actor (policy-as-code).
+
+§15F THREE CANARY MEANINGS — deliberately NOT one generic flag:
+
+    READINESS  — may this stage begin?   failure BLOCKS the stage
+    LIVENESS   — is progress still possible?  inactivity ≠ completion
+    EXIT       — did reality reach the expected state, invariants held?
+                 a stage cannot become PASS without it
+
+§15F INDEPENDENT COMPLETION AUTHORITY (14X.8 continued): the actor that
+performed the stage action may NOT self-certify its exit canary. The
+canary executes the check itself (command, readback, assertion) and
+records evidence; actor claims are evidence INPUT only.
+
+§15G TEST TRUTH: exit code 0 is not proof. Where the contract requires
+tests, the exit canary requires: non-zero executed test count, no
+skip-all, expected artifact present. Expectations come from the
+contract, never hard-coded.
+
+§15H HENXELS PLACEMENT: structural/system-health guard at proportionate
+boundaries — REQUIRED for structural integration boundaries and
+FULL/PROJECT_BUILD closure, NOT_APPLICABLE for tiny bounded changes.
+Every material boundary records which, with rationale.
+
+§15I DESIRED VS ACTUAL: after consequential actions — persist, reload
+canonical reality, compare against desired. A command exiting 0 is not
+an effect. Zero/acceptable delta gates the NEXT stage.
+"""
+
+from __future__ import annotations
+
+import re
+import subprocess
+import time
+from typing import Any
+from dataclasses import dataclass, field
+from pathlib import Path
+
+
+# ── §15F guarded stage contract ─────────────────────────────────────────────
+
+@dataclass
+class StageCheck:
+    """One named, executable readiness/exit check. The CHECK executes —
+    it is never satisfied by a claim."""
+    name: str
+    kind: str                  # "command" | "path_exists" | "assert"
+    # command: argv list; truth from exit code (+ optional output_pattern)
+    argv: list | None = None
+    output_pattern: str = ""   # regex that must appear on stdout (15G)
+    # path_exists: path that must exist after the stage
+    path: str = ""
+    # assert: callable(state) -> bool (in-process deterministic check)
+    check: Any = None
+    description: str = ""
+
+
+@dataclass
+class StageContract:
+    """Minimum guarded-stage semantics expressed over existing
+    construction state. Not a workflow engine: a stage contract is data
+    attached to (or derived from) a Construction Contract stage."""
+    stage_id: str
+    contract_id: str = ""
+    dependencies: list = field(default_factory=list)   # stage_ids
+    ready_when: list = field(default_factory=list)     # [StageCheck]
+    done_when: list = field(default_factory=list)      # [StageCheck]
+    invariants: list = field(default_factory=list)     # [StageCheck]
+    expected_effect: str = ""
+    retry_class: str = "UNKNOWN"   # TRANSIENT/INTERMITTENT/KNOWN_PROCESS_
+    #                              DEFECT/PERMANENT/APPROVAL_REQUIRED/UNKNOWN
+    idempotency_key: str = ""      # operation identity for null-action
+
+    def to_dict(self) -> dict:
+        def chk(c: StageCheck) -> dict:
+            return {"name": c.name, "kind": c.kind, "argv": c.argv,
+                    "output_pattern": c.output_pattern, "path": c.path,
+                    "description": c.description}
+        return {"stage_id": self.stage_id, "contract_id": self.contract_id,
+                "dependencies": list(self.dependencies),
+                "ready_when": [chk(c) for c in self.ready_when],
+                "done_when": [chk(c) for c in self.done_when],
+                "invariants": [chk(c) for c in self.invariants],
+                "expected_effect": self.expected_effect,
+                "retry_class": self.retry_class,
+                "idempotency_key": self.idempotency_key}
+
+
+class StageBlocked(Exception):
+    def __init__(self, stage_id: str, failures: list[dict]):
+        self.stage_id = stage_id
+        self.failures = failures
+        super().__init__(f"stage {stage_id} blocked: "
+                         f"{[f['name'] for f in failures]}")
+
+
+class ExitCanaryFailed(Exception):
+    def __init__(self, stage_id: str, failures: list[dict]):
+        self.stage_id = stage_id
+        self.failures = failures
+        super().__init__(f"stage {stage_id} exit canary FAILED: "
+                         f"{[f['name'] for f in failures]}")
+
+
+def _run_check(check: StageCheck) -> dict:
+    """Execute ONE check and record evidence. Returns
+    {name, passed, evidence}. A check that cannot execute is a FAILURE
+    (cannot-execute is never a pass)."""
+    ev: dict = {"name": check.name, "kind": check.kind}
+    try:
+        if check.kind == "command":
+            r = subprocess.run(check.argv or [], capture_output=True,
+                               text=True, timeout=300)
+            ev["exit_code"] = r.returncode
+            passed = r.returncode == 0
+            if passed and check.output_pattern:
+                m = re.search(check.output_pattern, r.stdout or "")
+                ev["pattern"] = check.output_pattern
+                ev["matched"] = bool(m)
+                passed = bool(m)
+                if m:
+                    ev["matched_text"] = m.group(0)[:100]
+        elif check.kind == "path_exists":
+            p = Path(check.path)
+            ev["path"] = check.path
+            passed = p.exists() and p.stat().st_size >= 0
+            ev["size"] = p.stat().st_size if p.exists() else 0
+        elif check.kind == "assert":
+            ev["assert"] = check.description or check.name
+            passed = bool(check.check({}))
+        else:
+            ev["error"] = f"unknown check kind {check.kind}"
+            passed = False
+    except Exception as e:  # noqa: BLE001 — cannot-execute is a failure
+        ev["error"] = str(e)[:200]
+        passed = False
+    ev["passed"] = passed
+    return ev
+
+
+def readiness_canary(stage: StageContract) -> dict:
+    """§15F: may this stage begin? ALL ready_when checks must pass.
+    Failure BLOCKS the stage — it does NOT auto-repair."""
+    failures = [ev for ev in (_run_check(c) for c in stage.ready_when)
+                if not ev["passed"]]
+    verdict = "READY" if not failures else "BLOCKED"
+    return {"stage_id": stage.stage_id, "canary": "readiness",
+            "verdict": verdict,
+            "checks": len(stage.ready_when), "failures": failures}
+
+
+def execute_stage(stage: StageContract, action) -> dict:
+    """Run one guarded stage: readiness → action → exit canary.
+    The action is the caller's function; its return value is evidence
+    INPUT. The exit canary independently verifies done_when + invariants.
+    Readiness failure raises StageBlocked (stage does not execute);
+    exit failure raises ExitCanaryFailed (stage cannot become PASS)."""
+    ready = readiness_canary(stage)
+    if ready["verdict"] == "BLOCKED":
+        raise StageBlocked(stage.stage_id, ready["failures"])
+    action_evidence = action() if action else {}
+    exit_evidence = [_run_check(c) for c in stage.done_when]
+    invariant_evidence = [_run_check(c) for c in stage.invariants]
+    failures = [e for e in exit_evidence + invariant_evidence
+                if not e["passed"]]
+    result = {"stage_id": stage.stage_id, "canary": "exit",
+              "verdict": "PASS" if not failures else "FAILED",
+              "action_evidence": action_evidence,
+              "exit_checks": exit_evidence,
+              "invariant_checks": invariant_evidence,
+              "failures": failures}
+    if failures:
+        # §15F: the stage cannot become PASS — and "cannot" means the
+        # caller gets an exception, not a verdict it may ignore.
+        raise ExitCanaryFailed(stage.stage_id, failures)
+    return result
+
+
+def liveness_canary(last_progress_at: float, now: float | None = None,
+                    stall_limit_secs: float = 300.0) -> dict:
+    """§15F: is progress still possible? Inactivity NEVER means
+    completion — a stalled stage is STALLED, not done."""
+    now = now if now is not None else time.monotonic()
+    stalled = (now - last_progress_at) > stall_limit_secs
+    return {"canary": "liveness", "stalled": stalled,
+            "idle_secs": round(now - last_progress_at, 1),
+            "verdict": "STALLED" if stalled else "ALIVE"}
+
+
+# ── §15G verify test truth, not just exit code ──────────────────────────────
+
+@dataclass
+class TestExpectations:
+    """Contract-derived verification expectations (never universal
+    constants)."""
+    require_tests: bool = True
+    min_tests: int = 1
+    test_output_pattern: str = r"(?:# pass (\d+)|(\d+) passed)"
+    forbid_all_skipped: str = r"# skipped (\d+)"
+    required_artifacts: list = field(default_factory=list)  # paths
+
+
+def verify_test_truth(cmd: list, cwd: str, expectations: TestExpectations,
+                      *, base_dir: str = ".") -> dict:
+    """§15G: run the verification command and judge TRUTH, not exit code.
+    False-green hostile cases rejected: zero tests discovered with exit 0,
+    all tests skipped, expected artifact absent."""
+    r = subprocess.run(cmd, cwd=cwd, capture_output=True, text=True,
+                       timeout=300)
+    evidence: dict = {"cmd": cmd, "exit_code": r.returncode,
+                      "stdout_tail": (r.stdout or "")[-500:]}
+    failures: list[dict] = []
+    if r.returncode != 0:
+        failures.append({"name": "exit_code", "exit_code": r.returncode})
+    if expectations.require_tests:
+        m = re.search(expectations.test_output_pattern, r.stdout or "")
+        count = int(m.group(1) or m.group(2)) if m else 0
+        evidence["test_count"] = count
+        if count == 0:
+            # zero tests + exit 0 is a false green when tests are required
+            failures.append({"name": "test_count", "count": 0,
+                             "why": "zero tests executed — false green"})
+        elif count < expectations.min_tests:
+            failures.append({"name": "test_count", "count": count,
+                             "min_required": expectations.min_tests})
+        if expectations.forbid_all_skipped:
+            s = re.search(expectations.forbid_all_skipped, r.stdout or "")
+            if s and count and int(s.group(1)) >= count:
+                failures.append({"name": "all_skipped",
+                                 "skipped": int(s.group(1))})
+    for art in expectations.required_artifacts:
+        p = Path(base_dir) / art
+        if not p.exists() or p.stat().st_size == 0:
+            failures.append({"name": "artifact_missing", "path": art})
+    evidence["failures"] = failures
+    evidence["verdict"] = "PASS" if not failures else "FALSE_GREEN_REFUSED"
+    return evidence
+
+
+# ── §15H Henxels as proportional structural guard ───────────────────────────
+
+def henxels_placement(*, depth: str, structural_boundary: bool,
+                      stage_id: str) -> dict:
+    """§15H: is Henxels REQUIRED at this boundary? Proportional, not
+    ceremonial. Records the decision with rationale (either way)."""
+    if structural_boundary or depth in ("FULL", "PROJECT_BUILD"):
+        return {"stage_id": stage_id, "henxels": "HENXELS_REQUIRED",
+                "rationale": (f"depth={depth}, structural_boundary="
+                              f"{structural_boundary} — structural "
+                              f"integration evidence required")}
+    return {"stage_id": stage_id, "henxels": "HENXELS_NOT_APPLICABLE",
+            "rationale": (f"depth={depth}, no structural boundary — "
+                          f"focused exit canary suffices")}
+
+
+# ── §15I desired vs actual reconciliation ───────────────────────────────────
+
+def reconcile(desired: dict, read_actual, *, stage_id: str = "",
+              tolerance: set | None = None) -> dict:
+    """§15I: persist → reload canonical reality → compare. `read_actual`
+    is a callable returning the ACTUAL canonical state (fresh read, never
+    a cached copy). Returns the delta; empty delta = continue. The next
+    stage MUST NOT start on a non-empty delta."""
+    tolerance = tolerance or set()
+    actual = read_actual()
+    delta: dict = {}
+    for key, want in desired.items():
+        if key in tolerance:
+            continue
+        got = actual.get(key) if isinstance(actual, dict) else None
+        if got != want:
+            delta[key] = {"desired": want, "actual": got}
+    return {"stage_id": stage_id, "reconciled": not delta, "delta": delta}

store/evidence/t15/README.md

@@ -0,0 +1,32 @@
+# Task Ledger
+
+A small standalone task ledger backed by SQLite (stdlib `sqlite3` only).
+Supports adding tasks, listing them, and marking them complete.
+
+## Files
+
+- `ledger.py` — the ledger (`connect`, `add_task`, `list_tasks`,
+  `complete_task`; SQLite storage, `:memory:` by default or a file path).
+- `ledger.test.py` — stdlib `unittest` suite proving add, list and
+  complete work, including SQLite file persistence.
+
+## Usage
+
+```python
+import ledger
+
+conn = ledger.connect("tasks.db")   # or ledger.connect() for in-memory
+task = ledger.add_task(conn, "write report")
+print(ledger.list_tasks(conn))
+ledger.complete_task(conn, task["id"])
+```
+
+## Run tests
+
+The test file uses a Node-style name (`ledger.test.py`), so the default
+`unittest discover` pattern (`test*.py`) does not match it. Run it
+directly:
+
+```sh
+python3 ledger.test.py -v
+```

store/evidence/t15/ledger.py

@@ -0,0 +1,94 @@
+"""Standalone SQLite task ledger: add, list and complete tasks.
+
+Stdlib only (sqlite3). State lives in SQLite; the caller owns the
+connection (``:memory:`` by default, or a file path for persistence).
+"""
+
+from __future__ import annotations
+
+import sqlite3
+
+SCHEMA = (
+    "CREATE TABLE IF NOT EXISTS tasks ("
+    "id INTEGER PRIMARY KEY AUTOINCREMENT, "
+    "title TEXT NOT NULL, "
+    "completed INTEGER NOT NULL DEFAULT 0"
+    ")"
+)
+
+
+def connect(path: str = ":memory:") -> sqlite3.Connection:
+    """Open a ledger database and ensure the schema exists."""
+    conn = sqlite3.connect(path)
+    conn.row_factory = sqlite3.Row
+    init_db(conn)
+    return conn
+
+
+def init_db(conn: sqlite3.Connection) -> None:
+    """Create the tasks table if it does not exist."""
+    conn.execute(SCHEMA)
+    conn.commit()
+
+
+def _row_to_dict(row: sqlite3.Row) -> dict:
+    return {
+        "id": row["id"],
+        "title": row["title"],
+        "completed": bool(row["completed"]),
+    }
+
+
+def add_task(conn: sqlite3.Connection, title: str) -> dict:
+    """Add a task; return ``{"id", "title", "completed"}``.
+
+    Raises:
+        TypeError: if title is not a string.
+        ValueError: if title is empty or blank.
+    """
+    if not isinstance(title, str):
+        raise TypeError("title must be a string")
+    cleaned = title.strip()
+    if not cleaned:
+        raise ValueError("title must be a non-empty string")
+    init_db(conn)
+    cur = conn.execute(
+        "INSERT INTO tasks (title, completed) VALUES (?, 0)", (cleaned,)
+    )
+    conn.commit()
+    task_id = cur.lastrowid
+    row = conn.execute(
+        "SELECT id, title, completed FROM tasks WHERE id = ?", (task_id,)
+    ).fetchone()
+    return _row_to_dict(row)
+
+
+def list_tasks(conn: sqlite3.Connection) -> list:
+    """Return all tasks ordered by id (each a dict)."""
+    init_db(conn)
+    rows = conn.execute(
+        "SELECT id, title, completed FROM tasks ORDER BY id"
+    ).fetchall()
+    return [_row_to_dict(r) for r in rows]
+
+
+def complete_task(conn: sqlite3.Connection, task_id: int) -> dict:
+    """Mark the task with ``task_id`` completed; return the updated dict.
+
+    Raises:
+        TypeError: if task_id is not an int.
+        LookupError: if no task has that id.
+    """
+    if isinstance(task_id, bool) or not isinstance(task_id, int):
+        raise TypeError("task_id must be an int")
+    init_db(conn)
+    cur = conn.execute(
+        "UPDATE tasks SET completed = 1 WHERE id = ?", (task_id,)
+    )
+    conn.commit()
+    if cur.rowcount == 0:
+        raise LookupError(f"no task with id {task_id}")
+    row = conn.execute(
+        "SELECT id, title, completed FROM tasks WHERE id = ?", (task_id,)
+    ).fetchone()
+    return _row_to_dict(row)

store/evidence/t15/ledger.test.py

@@ -0,0 +1,136 @@
+"""Tests for the standalone SQLite task ledger (``ledger.py``).
+
+Run directly (the file name is Node-style, so the default
+``python -m unittest discover`` pattern ``test*.py`` does not pick it up)::
+
+    python3 ledger.test.py -v
+
+Stdlib only.
+"""
+
+import os
+import sqlite3
+import tempfile
+import unittest
+
+import ledger
+
+
+def fresh_conn():
+    return ledger.connect(":memory:")
+
+
+class TestAddTask(unittest.TestCase):
+    def test_add_returns_task_with_id_title_and_not_completed(self):
+        conn = fresh_conn()
+        task = ledger.add_task(conn, "write report")
+        self.assertEqual(task["title"], "write report")
+        self.assertFalse(task["completed"])
+        self.assertIsInstance(task["id"], int)
+
+    def test_add_assigns_unique_ids(self):
+        conn = fresh_conn()
+        first = ledger.add_task(conn, "one")
+        second = ledger.add_task(conn, "two")
+        self.assertNotEqual(first["id"], second["id"])
+
+    def test_add_trims_title(self):
+        conn = fresh_conn()
+        task = ledger.add_task(conn, "  tidy desk  ")
+        self.assertEqual(task["title"], "tidy desk")
+
+    def test_add_rejects_empty_or_blank_title(self):
+        conn = fresh_conn()
+        with self.assertRaises(ValueError):
+            ledger.add_task(conn, "")
+        with self.assertRaises(ValueError):
+            ledger.add_task(conn, "   ")
+
+    def test_add_rejects_non_string_title(self):
+        conn = fresh_conn()
+        with self.assertRaises(TypeError):
+            ledger.add_task(conn, None)
+        with self.assertRaises(TypeError):
+            ledger.add_task(conn, 123)
+
+
+class TestListTasks(unittest.TestCase):
+    def test_list_empty_ledger(self):
+        self.assertEqual(ledger.list_tasks(fresh_conn()), [])
+
+    def test_list_returns_all_tasks_in_order(self):
+        conn = fresh_conn()
+        ledger.add_task(conn, "alpha")
+        ledger.add_task(conn, "beta")
+        tasks = ledger.list_tasks(conn)
+        self.assertEqual([t["title"] for t in tasks], ["alpha", "beta"])
+
+    def test_list_shows_completed_flags(self):
+        conn = fresh_conn()
+        task = ledger.add_task(conn, "alpha")
+        ledger.add_task(conn, "beta")
+        ledger.complete_task(conn, task["id"])
+        by_title = {t["title"]: t for t in ledger.list_tasks(conn)}
+        self.assertTrue(by_title["alpha"]["completed"])
+        self.assertFalse(by_title["beta"]["completed"])
+
+
+class TestCompleteTask(unittest.TestCase):
+    def test_complete_marks_matching_task(self):
+        conn = fresh_conn()
+        first = ledger.add_task(conn, "alpha")
+        second = ledger.add_task(conn, "beta")
+        updated = ledger.complete_task(conn, first["id"])
+        self.assertTrue(updated["completed"])
+        self.assertEqual(updated["id"], first["id"])
+        by_title = {t["title"]: t for t in ledger.list_tasks(conn)}
+        self.assertTrue(by_title["alpha"]["completed"])
+        self.assertFalse(by_title["beta"]["completed"])
+
+    def test_complete_persists_across_listing(self):
+        conn = fresh_conn()
+        task = ledger.add_task(conn, "alpha")
+        ledger.complete_task(conn, task["id"])
+        done = [t for t in ledger.list_tasks(conn) if t["completed"]]
+        self.assertEqual([t["title"] for t in done], ["alpha"])
+
+    def test_complete_unknown_id_raises(self):
+        conn = fresh_conn()
+        ledger.add_task(conn, "alpha")
+        with self.assertRaises(LookupError):
+            ledger.complete_task(conn, 999)
+
+    def test_complete_rejects_non_int_id(self):
+        conn = fresh_conn()
+        ledger.add_task(conn, "alpha")
+        with self.assertRaises(TypeError):
+            ledger.complete_task(conn, "1")
+
+
+class TestPersistence(unittest.TestCase):
+    def test_tasks_survive_reconnect_to_file_db(self):
+        with tempfile.TemporaryDirectory() as tmp:
+            path = os.path.join(tmp, "tasks.db")
+            conn = ledger.connect(path)
+            task = ledger.add_task(conn, "persist me")
+            ledger.complete_task(conn, task["id"])
+            conn.close()
+            reopened = ledger.connect(path)
+            try:
+                tasks = ledger.list_tasks(reopened)
+            finally:
+                reopened.close()
+            self.assertEqual(len(tasks), 1)
+            self.assertEqual(tasks[0]["title"], "persist me")
+            self.assertTrue(tasks[0]["completed"])
+
+    def test_uses_sqlite_storage(self):
+        conn = fresh_conn()
+        self.assertIsInstance(conn, sqlite3.Connection)
+        ledger.add_task(conn, "alpha")
+        rows = conn.execute("SELECT title FROM tasks").fetchall()
+        self.assertEqual([r[0] for r in rows], ["alpha"])
+
+
+if __name__ == "__main__":
+    unittest.main(verbosity=2)

store/evidence/t15/run-q-tests.txt

@@ -0,0 +1,2 @@
+# tests 10
+# pass 10

store/evidence/t16/16A_baseline.md

@@ -0,0 +1,48 @@
+# T16 Stage 16A — exact baseline (verified 2026-09-15 ~15:40 local)
+
+Fresh session ses_-ffe5f5a82d0b5ffeLrGt5rMRv. All facts verified independently this session; none carried from prior-session memory without re-verification.
+
+## Identity / release chain
+- gitea/master: 5bd8cca9410b = auto-bump 0.66.62 (2026-09-15 14:15:22 +0100)
+- Releases: v0.66.62 published 14:15:51; v0.66.60 (13:29), v0.66.61 (13:42) same day
+- Deployed: /mnt/workdrive/rune-code/current -> 5bd8cca9410b (symlink mtime 14:20)
+- last-good -> fd5f31b (0.66.61-lineage fix commit)
+- CONVERGED: release tag, deployed current, and master tip are the same SHA. No manual refresh used.
+
+## Inherited T15 open item — RESOLVED on master
+- 15Z CI failure (196cf19: pyright 445 vs 443; unit-tests job truncated) — fixed by:
+  - 193d819 "event_wait importlib-clean loading, zero pyright warnings" (= be87dc3 rebased/pushed)
+  - b2d6802 "§15K test no longer shells out to git worktree on CI"
+  - fd5f31b "§15D tests use ephemeral ports, not fixed 8793"
+- These drove releases 0.66.60 → 0.66.61 → 0.66.62, each deployed. Nothing inherited is left open.
+
+## Working copies
+- /mnt/workdrive/issue161-harness: master tip, clean — THE code worktree for T16.
+- /mnt/workdrive/rune (RUNE_HOME): branch pinned-live @ 90c7db1 (2026-09-08), 652 commits behind master, HEAD is a clean ancestor (fast-forwardable). Expected: RUNE_HOME is the data-continuity home; runtime launches from rune-code/current. Still: stale and drifting (16P material).
+- 39 stale build-20260915-* worktrees, all cut at 90c7db1 (stale base) — /home/nick/.rune/worktrees/. Cleanup candidate (16W).
+
+## Gates (all run in lock-exact environments, master tip 5bd8cca)
+- Full suite: 1854 passed + 84 subtests, 0 failures (113s)
+- Lint ratchet: PASS (BLE001 354/354, S110 80/80 at baseline). Raw `ruff check` = 12 errors, all within committed lint-baseline (mandate's "8 pre-existing" is now 12 — drift, 16W item).
+- Type-check ratchet: PASS — errors 117 (= baseline), warnings 437 (≤ 443). CRITICAL METHOD NOTE: must run in `.type-venv` recreated from requirements-ci.txt (lock-exact). Stale /tmp/ci-venv (pyright 1.1.414 but stale env) false-failed 124 vs 117. Environment-exactness is a repeat failure class (16U).
+- Henxels: rc=0.
+
+## Wait/resume infrastructure
+- W/R daemon RUNNING: pid 3756873, `python3 daemon.py 8791` (webhook receiver on 0.0.0.0:8791), restarted 14:08 today.
+- Daemon lives at /home/nick/.local/share/mimo-wait-resume (daemon.py, waitstore.py, verify.py, own tests) with its OWN git snapshot @ 4811070 "T15 §15D/§15E: per-wait resume_cmd override + PROCESS_EVENT wait kind + /process-event endpoint". The daemon is NOT part of the master tree — src/rune/ has event_wait.py (client-side adapter) but no daemon.py. (16M inventory finding: out-of-repo deployment, separate snapshot.)
+- Wait store: 45 records — states: COMPLETED 19, RESUME_UNVERIFIED 10, FAILED 6, READY_TO_RESUME 6, WAITING_EXTERNAL 4. Kinds: EXACT_WORKFLOW 24, RELEASE_DEPLOY 8, PROCESS_EVENT 7 (+6 pre-kind records with wait_kind absent).
+- 4 WAITING_EXTERNAL + 6 READY_TO_RESUME + 10 RESUME_UNVERIFIED = outstanding waits to reconcile or leave truthfully parked.
+
+## Scaffolding stores
+- Repo store/scaffold/current.json: spine-shaped (entities/relationships/observations), no projects key.
+- store/scaffold/projects/: EMPTY. T14/T15 project scaffolds lived in ephemeral /tmp stores (e.g. /tmp/t15r-store). Durable evidence: store/evidence/t14/, store/evidence/t15/ (+ t16/ created this session).
+
+## T15 maturity — CORRECTED RECORD (16A mandate)
+- Programme continuation MECHANISM (stage_guards, event waits, canaries): INTEGRATED / TEST-PROVEN. T15's own test evidence stands — not erased.
+- Uninterrupted REAL-SESSION programme continuation: NOT YET LIVE-PROVEN. The prior session's actual user experience was repeated manual "continue" despite the mechanism being tested green. This fresh T16 session is the live experiment (Stage 16B).
+- Builder event-driven wait: CANONICAL DEFAULT (bounded live proof, T15).
+- CI/deployment wait: STILL POLLING FALLBACK (sleep-N → query patterns remain in release/CI verification paths) — the 16C/D target. Verified by inspection during 16A setup; full inventory in 16C.
+
+## Provider/model handling (initial scan, full audit = 16F)
+- Known hardcode surface to classify: Verboo, glm-5.3-flash, mimo-v2.5-pro, Xioami, BUILDER_MODEL, DEFAULT_MODEL, MIMO_MODEL, --model/-m.
+- T14/T15 real-builder runs pinned "Verboo Code/glm-5.3-flash" at the DRIVER layer (t14_builder_session.py), not inside Scaffolding proper — classification pending 16F audit.

store/evidence/t16/16B_continuity_log.md

@@ -0,0 +1,44 @@
+# T16 Stage 16B — live continuation diagnostic (spans the whole tranche)
+
+fresh_session = true
+session = ses_-ffe5f5a82d0b5ffeLrGt5rMRv
+tranche_start = 2026-09-15 (T16 mandate received at session open)
+
+## Response-boundary ledger
+
+| # | boundary | last completed | next eligible stage started automatically? | user intervention required? |
+|---|----------|----------------|---------------------------------------------|------------------------------|
+| 1 | session open (checkpoint rebuild from prior-session T16 mandate) | none | 16A started this session without user prompt | NO |
+
+## Stage log
+
+- 16A: STARTED 2026-09-15 (this session, unprompted continuation from mandate). COMPLETED ~15:45 local. Baseline + T15 maturity correction persisted (store/evidence/t16/16A_baseline.md + project MEMORY.md). Next stage (16C) selected and started immediately in the same assistant turn — no user intervention.
+
+## Verdict so far
+
+Running experiment. Required live proof: ≥4 substantial sequential stages completing and the next starting automatically without user intervention. A LIVE_CONTINUATION_FAILURE is recorded the moment this assistant reaches a normal response boundary while eligible authorized work remains.
+
+## Stage log update (16:00Z)
+
+- 16A: COMPLETED (persisted + readback done). No user intervention.
+- 16B diagnostic: created and maintained continuously. No user intervention.
+- 16C: COMPLETED (event-driven CI wait adapter implemented, 8 held-outs, workflows wired, daemon passthrough landed). NO user intervention — began immediately after 16A in the same assistant turn.
+- 16D: IN PROGRESS — first live cycle exposed a real defect (runner user vs bridge data root; deploy failed fast+truthfully on 7953/7958 — no false green). Root-caused, shared root + readiness probe shipped (c707a81, pushed 16:02Z); latency proof pending the next cycle. This is itself a 16U datum: "wrong environment" was caught by production before a canary could — the readiness probe now added prevents recurrence.
+- 16E: COMPLETED (one primitive: builder + gitea adapters share register/FIFO/match/idempotency/watchdog; identity selects events — 16E held-out).
+- 16F/16G: COMPLETED (provider_context seam, PROVIDER_UNAVAILABLE class, 7 silent provider defaults removed, structural guard + 6 held-outs; committed 37deb60).
+
+Response boundaries so far: ZERO user interventions. Continuation has been automatic at every stage boundary.
+
+## Stage log update (16:35Z)
+- 16D: COMPLETED — two real production cycles proved the watchdog/
+  deadline truthfulness path (deploy promoted from the deadline path,
+  zero false greens, sleep-30 floor gone); event-before-waiter root-
+  caused and fixed via durable replay (b51afea pushed 16:34Z).
+- 16E: COMPLETED (one primitive, adapters by identity).
+- 16F/16G: COMPLETED. 16I/J/K: COMPLETED. 16M: COMPLETED. 16N: COMPLETED
+  (docs/STRUCTURAL_MAP.md). 16O: COMPLETED (honest result — no step-count
+  win on tiny fixtures; qualitative advantages recorded). 16P: COMPLETED
+  (drift_resume.py + real-fixture held-outs).
+- Next: 16Q (reconciliation), 16R (self-hosted change), 16U-V, 16W-Y, 16Z.
+- USER INTERVENTIONS TO DATE: ZERO. Every stage transition has been
+  automatic within continuous assistant execution.

store/evidence/t16/16D_latency_proof.md

@@ -0,0 +1,44 @@
+# T16 §16D — live CI/deploy latency proofs (2026-09-15)
+
+## Cycle 1: c707a81 (fix commit) — watchdog/deadline-recovery path
+- registered_at: 16:03:16Z (deploy job start)
+- event_received_at: (none — ci.yml notify ran BEFORE deploy registered;
+  event-before-waiter ordering, the workflow-level sibling of 16Y #10)
+- controller behavior: watchdog expiry → ci_alive() True (runs genuinely
+  non-terminal) → truthful re-arms; no false green at any point
+- verdict_at: 16:18:16Z (mode deadline_pending → final authority read
+  success), wall_secs 900.92 — the deadline is a FAILURE DETECTOR; it
+  expired exactly once and the final one-shot authority read decided
+- PROMOTED: c707a81 at 16:18:17Z (current moved; last-good → previous)
+- Sleep-30 polling floor: GONE (old loop would have polled 30 times in
+  this window; new adapter slept in select(), zero poll queries — the
+  only Gitea reads were one-shot authority verdicts on expiry)
+
+## Cycle 2: 4a06d9f (auto-bump 0.66.64) — true event path
+- waiter registered while CI still in progress; process observed
+  sleeping in select() (S state, 700s elapsed, zero CPU) at inspection
+- expected: notify-event-bridge POST on CI completion → mid-wait wake →
+  resume latency « 30s floor (proof line appended when the cycle closes)
+
+## Event-before-waiter finding (new, honest)
+The wake event cannot be delivered when CI finishes before the deploy
+job registers its correlation. This ordering is inherent (deploy.yml
+starts after checkout; ci.yml may finish earlier). The durable-bridge
+recovery covers it: the waiter's watchdog + one-shot authority read.
+A future improvement could register the correlation BEFORE CI completes
+(e.g. a preceding workflow step) — deferred; watchdog recovery is
+truthful and the deadline bound held exactly.
+
+## Cycle 2 postmortem + replay fix (17:35Z)
+Cycle 2 (4a06d9f) also ended deadline_pending at 901.0s: the completion
+events DID reach the daemon (16:07:44/16:08:21, after registration at
+16:07:06) but never resumed the waiter. ROOT CAUSE: on a serialized
+single-runner machine the deploy job starts only after ci.yml's job
+slot frees — the wake lands before/while the waiter is between
+registration arms; the FIFO rendezvous is forward-only.
+FIX: durable event replay — find_matching_event() scans the bridge's
+event log at registration; a matching event short-circuits straight to
+the authority read (event_replayed mode). 2 held-outs added (9 total).
+NOTE: the replay fix lands on the NEXT cycle (it cannot prove itself on
+the cycle that exposed it); cycles 1-2 remain valid proof of the
+watchdog/deadline truthfulness path.

store/evidence/t16/16M_inventory.md

@@ -0,0 +1,54 @@
+# T16 Stage 16M — Retroactive Rune inventory (2026-09-15)
+
+Sources: two parallel read-only exploration sweeps + 16A/16C hands-on verification. Classification scale per mandate. This is the raw material for the 16N structural map.
+
+## The ten canonical systems
+
+| System | Modules | Canonical state | Classification | Notes |
+|---|---|---|---|---|
+| Spine | spine.py (695L) over structural_store.py (650L, zero Rune constants) | store/spine/current.json, snapshots byte-deterministic | HEALTHY | maturity ladder evidence-gated; CLI JSON-canonical |
+| Research pipeline | scripts/research_task_tree.py (authority), research_stage_executor.py, research_event_relay.py | research/issue-N/ artifacts + Gitea labels (labels=projection) | HEALTHY | stage gates per artifact; coverage thinner than Spine |
+| ADAPT/router | adaptive_router.py (625L, "canonical route authority ADAPT-2"), route_risk_flags.py | routes/route.jsonl append-only | HEALTHY | capability_broker/recall_router/warm_router/model_routing are SEPARATE routing domains — same word, different authority; DO NOT MERGE |
+| Completion Truth | verification_layer.py, orchestration/conversation_state.py (TaskStateRepository over DurableEventStore), records.py | event-store log + atomic checkpoints | UNCLEAR_AUTHORITY / MISSING_SEAM | no enforced seam that a task cannot close COMPLETED while verifiers fail — highest-value gap found |
+| Release/deploy | deploy_stage/deploy_promotion/deploy_qualify/create_release/auto_bump + workflows | deploy root + current/last-good + _build_identity.json | HEALTHY | monotonic promotion fail-closed; 16C event-driven wait now added |
+| Provider layer | model_routing ("one authoritative router #29 P1-06"), model_connect, model_registry, provider_registry, work_manager, standardized_providers, openai_compat_transport | model_routing_profiles.json keyed (provider,model) | DUPLICATE_ROUTE / UNCLEAR_AUTHORITY | ≥5 selection entry points; the OPERATOR is the final decider, not any registry. 16F removed construction-side selection; provider-layer consolidation = 16Q candidate |
+| W/R event bridge | event_wait.py + out-of-repo daemon.py/waitstore.py/verify.py | WR_ROOT waits/events/leases/resumes | HEALTHY / OPTIONAL_BUT_SHOULD_BE_DEFAULT | one primitive, builder+gitea adapters (16E proven); daemon lives OUT-OF-REPO with its own git snapshot — packaging/deployment debt |
+| TUI/backend | rune_tui.py (3885L) ← rune_operator.py (2604L) + microkernel + orchestration | conversation/task repositories | HEALTHY | one-way direction enforced by canary_tui_drift; import via sys.path shim (fragile, keep test) |
+| build_mode (legacy /build) | build_mode.py BuildManager + plugins/build_mode/phases | ~/.rune/work/builds + worktrees | LEGACY / DUPLICATE_ROUTE | superseded by builder_bridge + event-driven sessions (T14/T15); still wired → two ways to run a build; REPO_DIR="/mnt/workdrive/rune" stale constant |
+| Canaries | 24 canary_*.py + canary_native.MODULES registry | none (stateless) | HEALTHY | native/full_system/orchestration_foundation partially overlap (layered, acceptable); no test enforces MODULES completeness |
+
+## Auxiliary systems
+
+| System | Classification | Notes |
+|---|---|---|
+| Scaffolding itself | HEALTHY | provider-blind since 16F (structural guard); phase machine governed |
+| Maintenance machinery | UNCLEAR_AUTHORITY | three overlapping routes: scaffold maintenance_item entities, tui_backlog ledger, drift_ledger — no central module, NO tests |
+| Task/intent state | DUPLICATE_ROUTE | task-tool session DB vs task_persist persisted.json vs memory_store.working vs orchestration event store — four task-ish stores |
+| Henxels | HEALTHY | henxels.yaml + 191L checks; contract-tested |
+| Plugin surfaces | MISSING_SEAM | phases exist as files; no generic loader; single hardcoded plugin |
+| Learning/process records | DUPLICATE_ROUTE | memory_store vs graph_memory vs warm_neuron_store vs curator vs process_experience — no single authority doc |
+| programme_continuation | MISSING_SEAM | run_programme() has NO production caller (proof-only, T15 §15M); night_mode LEGACY aspirational; cron_engine MISSING_VERIFICATION |
+| maintenance.yml | LEGACY | canary-suite job loops shared/canary_*.py — real canaries live in src/rune/; the job likely no-ops |
+
+## Duplicate routes / bypasses (16Q raw material, priority-ordered)
+
+1. **task-state quadruplication** — 4 stores hold task-ish state; truth hierarchy unclear
+2. **build_mode vs builder_bridge** — two build paths, legacy still wired
+3. **provider selection ×5 entry points** — operator is de-facto authority
+4. **T14 vs T15 builder drivers** — T15 is the event-driven successor; T14 parallel-runs
+5. **maintenance ledgers ×3** — scaffold entities, backlog, drift ledger
+6. **maintenance.yml canary glob wrong dir** — stale no-op job
+7. **programme_continuation unwired** — the "continue problem" machinery exists but nothing calls it in production
+8. **memory stores ×5** — overlapping recall surfaces
+
+## Hard-coded dev paths in production code (16W targets)
+
+- build_mode.REPO_DIR + 5 complexity_scorer defaults → /mnt/workdrive/rune (stale for harness)
+- deploy_promotion.DEPLOY_ROOT → /mnt/workdrive/rune-code (fine: that IS the deploy root, but should be env-derived — it is)
+- two canaries sys.path.insert(/home/nick/.config/opencode/scripts)
+- memory/system_registry.json points at /mnt/workdrive/rune — STALE repo root
+- event_wait FALLBACK_WR_ROOT → /home/nick/... (fallback only; shared root now env-set in deploy.yml)
+
+## Honest overall
+
+Rune's structural health is better than the mandate's tone suggests: Spine, release/deploy, ADAPT, W/R, TUI boundary, canaries, Henxels, and Scaffolding are all HEALTHY with real verification. The debt concentrates in: completion-truth enforcement seam, task-state multiplication, provider-layer entry-point spread, legacy build_mode still wired, and unwired programme continuation (the very continuation defect this tranche experiments on).

store/evidence/t16/16O_benchmark.md

@@ -0,0 +1,45 @@
+# T16 §16O — long-gap re-entry benchmark (2026-09-15)
+
+Fixture: /mnt/workdrive/t16-reentry — Python invoice-ledger project, 2
+stages (1 closed+verified, 2 blocked PROVIDER_UNAVAILABLE with recorded
+next action), then POST-CHECKPOINT DRIFT injected: (a) VAT regression in
+report.py breaking 2 tests, (b) unrecorded discount.py module, (c) 2
+unrecorded commits, (d) committed __pycache__ + no .gitignore (unplanned
+extras). Two fresh-identity subagents, no transcript, no briefing beyond
+method.
+
+## Measured
+
+| Metric | PATH A cold rediscovery | PATH B scaffold-assisted |
+|---|---|---|
+| Files opened | 8 | 9 |
+| Commands run | 4 | 6 |
+| Explore/verify cycles | ~5 | ~8 |
+| Drifts found | 4/4 planted + pycache + gitignore | 4/4 planted (+pytest_cache receipt); did not enumerate pycache/gitignore |
+| False assumptions corrected | 2 (pytest path; comments-might-be-stale) | 3 (pytest path; assumed green still held; assumed repo at checkpoint commit) |
+| Drift CLASSIFICATION | none (flat findings) | EXPECTED/MATERIAL applied correctly, per-claim CONFIRMED/DRIFTED |
+| Next action quality | correct, self-derived | correct + explicitly adjusted from recorded next action |
+| Confidence calibration | honest (blocked=medium) | honest (drift completeness medium-high) |
+
+## Honest verdict
+
+1. On a TINY fixture, cold rediscovery is CHEAP and nearly as accurate —
+   the scaffold did not save significant effort at this scale. No
+   fabricated win: file/command counts are comparable (8 vs 9, 4 vs 6).
+2. The scaffold's measurable advantages are QUALITATIVE and scale with
+   project size: (a) structured per-claim reconciliation instead of flat
+   discovery, (b) recorded next action + blocker evidence that survives
+   model turnover, (c) drift CLASSES enabling the 16P material/expected
+   gating decision, (d) intent (non_goals: "no hardcoded FX rates")
+   that cold inspection cannot infer.
+3. Both paths needed a test run to catch the regression — checkpoint
+   claims NEVER substitute for reality (the 16P principle, proven by
+   both paths independently).
+4. Path A's unplanned extras (pycache/gitignore) show cold discovery
+   also finds things outside the scaffold's frame — the scaffold is a
+   starting point, not a boundary.
+
+Recommendation: adopt scaffold-assisted re-entry as the DEFAULT for
+multi-session projects (16P flow), keep cold verification of claims as
+a mandatory reconciliation step. Re-benchmark on a realistically-sized
+repo before claiming step-efficiency wins (OVERFLOW 3 candidate).

store/evidence/t16/16STUVXY_findings.md

@@ -0,0 +1,91 @@
+# T16 §16S/§16T/§16U/§16V/§16X/§16Y — consolidated findings (2026-09-15)
+
+## §16S — live programme continuation proof
+Evidence: 16B_continuity_log.md. Stages 16A→16B→16C→16D→16E→16F/16G→
+16I/J/K→16M→16N→16O→16P→16Q/16R executed in ONE continuous assistant
+session with ZERO user "continue" interventions; every stage transition
+was automatic. ≥4 substantial sequential stages completed with the next
+starting automatically (requirement exceeded: 12+ stages).
+**LIVE-PROVEN (bounded, this session).** Caveat recorded honestly: the
+session is long; a context-boundary failure would still be possible in
+even longer programmes — the mechanism (stage_guards + programme_
+continuation canary) is what carried continuation, now live-proven.
+
+## §16T — session-freshness finding
+Previous long session (T15): user typed "continue" repeatedly DESPITE
+the mechanism being test-proven. Fresh T16 session: zero interventions
+across 12+ stages. Verdict: **MIXED** — old-session degradation is
+SUPPORTED as the parsimonious explanation (same mechanism, same model
+class, different session freshness → different behavior), but n=1 and
+the prior session also carried much larger context; a systemic harness
+contribution cannot be excluded. Not called corruption; no evidence of
+it.
+
+## §16U — interface-first error prevention (moved-left repairs)
+- pyright ambient-venv skew (117⇄124): root-caused; ratchet now
+  neutralizes VIRTUAL_ENV/PATH before invoking pyright — environment
+  mismatch is now detected-at-setup instead of post-hoc.
+- deploy readiness probe: RUNE_WR_ROOT writability checked BEFORE the
+  wait (the PermissionError class caught twice in production is now a
+  clear setup-time error).
+- provider PROVIDER_UNRESOLVED: missing provider context fails loudly
+  before any construction run instead of silently using a default.
+- lesson-class recorded: "wrong environment" failures were caught by
+  production twice before a canary existed — readiness-before-action is
+  the general fix shape.
+
+## §16V — canary + Henxels coverage vs Rune boundaries
+- Henxels rc=0 all session; umbrella manifest gate caught every new
+  module (provider_context, construction_projection, construction_
+  commands, drift_resume) — the completeness statement works.
+- Found + fixed: maintenance.yml canary glob pointed at shared/ (silent
+  no-op) — 16Q/16R.
+- Gaps recorded (16M): no MODULES-completeness test for canary registry;
+  maintenance machinery lacks tests; daemon out-of-repo has no CI cover.
+- Not duplicated: Henxels = structural statements; canaries = import/
+  exercise; tests = behavior. Boundary held.
+
+## §16X — backlogs (backend → future projection/control needs)
+Construction projection (16I) already exposes: project, contract, phase,
+stage, waiting_state, correlation, builder, blockers, verification,
+next_action. TUI/Substrate control needs mapped to 16J commands:
+START/RESUME/PROVIDE_APPROVAL/PROVIDE_CONTEXT + REQUEST_STATE.
+Drift display: drift_resume classes are the vocabulary.
+DEBT (recorded, not built): PAUSE needs a lifecycle phase if wanted;
+goal-level (why-is-this-active) linkage belongs to future Substrate.
+
+## §16Y — adversarial matrix (30 cases) — where each is held
+1 old-context absent → 16O both paths; 2 scaffold reconstructs → 16O Path B
++ 16P; 3 provider unavailable → 16G-C; 4 no fallback → 16G-D (policy 0);
+5 real status preserved → 16G-C verbatim; 6 duplicate event → 15E tests;
+7 stale event → 15E tests; 8 missing webhook → 16D cycles 1-2 (deadline
+recovery, promoted truthfully); 9 duplicate webhook → 15E double-resume
+tests; 10 event-before-waiter → REPLAY FIX (b51afea) + live finding;
+11 controller restart → 15E restart tests + resume_or_recover; 12 mock
+invalid command → 16K refusal tests; 13 direct-state semantics → 16K
+inexpressible test; 14 stale scaffold → 16P VAT regression blocks; 15
+months-later drift → 16O/16P; 16 incomplete work after gap → 16O stage-2;
+17 no-activity≠completion → watchdog/PROCESS_LOST semantics (15D/Q);
+18 duplicate route → 16M inventory + 16Q fix; 19-21 TUI request classes →
+classify_mode (T10-15 suites); 22 manual provider switch → 16G-G
+byte-identical scaffold; 23 approval gate → LEGAL_TRANSITIONS + 16J
+user-provenance refusal; 24 Henxels defect → 16Q canary glob (found+fixed
+by inventory); 25 false-green → stage_guards verify_test_truth (15F-I)
++ deploy refusals observed live; 26 completion replay → match only on
+WAITING_EXTERNAL; 27 stale publisher → T14B authority re-read; 28 CI
+completion event → 16C suite; 29 stage-completion-≠-programme → 15M
+canary + this session's behavior; 30 final completion → 16Z report
+condition (this document; programme-complete only when the release
+cycle converges).
+
+Gaps honestly recorded: adversarial cases 22-27 rely on T14/T15-era
+suites (not re-run live this tranche); cases 1-2 depend on subagent
+honesty (both paths cross-checked against planted ground truth).
+
+## §16Z — final release convergence (17:55Z)
+Deployed current -> 986b450 (T16 16O/P commit) at 17:54Z, promoted by
+the event-driven pipeline itself; releases 0.66.63→0.66.66 auto-published
+and deployed through T16 commits; zero manual release refresh. Run 7991
+(2a4f9c8) in flight at report time; durable waits remain parked and the
+pipeline drains autonomously. Suite 1884 green; ratchets PASS; Henxels
+rc=0 at last full gate run.

store/scaffold/programmes/t17-17e-proof-2.json

@@ -0,0 +1,36 @@
+{
+ "programme_id": "t17-17e-proof-2",
+ "objective": "17E live proof: 4 stages, async event wait, failure, auto-continuation",
+ "stages": [
+  {
+   "stage_id": "17e2-stage-1-governance",
+   "objective": "governance-only stage",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 0
+  },
+  {
+   "stage_id": "17e2-stage-2-async",
+   "objective": "async work under event bridge",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 1
+  },
+  {
+   "stage_id": "17e2-stage-3-fails",
+   "objective": "deliberately failing stage",
+   "state": "FAILED",
+   "depends_on": [],
+   "priority": 2
+  },
+  {
+   "stage_id": "17e2-stage-4-final",
+   "objective": "final stage",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 3
+  }
+ ],
+ "fallbacks": [],
+ "saved_at": "2026-09-15T17:51:22Z"
+}

store/scaffold/programmes/t17-17e-proof-3.json

@@ -0,0 +1,36 @@
+{
+ "programme_id": "t17-17e-proof-3",
+ "objective": "17E live proof: async event wait, failure isolation, auto-continuation, honest terminal verdict",
+ "stages": [
+  {
+   "stage_id": "17e3-stage-1-governance",
+   "objective": "governance-only stage",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 0
+  },
+  {
+   "stage_id": "17e3-stage-2-async",
+   "objective": "async work under event bridge",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 1
+  },
+  {
+   "stage_id": "17e3-stage-3-fails",
+   "objective": "deliberately failing stage",
+   "state": "FAILED",
+   "depends_on": [],
+   "priority": 2
+  },
+  {
+   "stage_id": "17e3-stage-4-final",
+   "objective": "final stage",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 3
+  }
+ ],
+ "fallbacks": [],
+ "saved_at": "2026-09-15T17:52:31Z"
+}

store/scaffold/programmes/t17-17e-proof.json

@@ -0,0 +1,36 @@
+{
+ "programme_id": "t17-17e-proof",
+ "objective": "prove canonical programme continuation: 4 stages incl. async event wait, failure, auto-continuation",
+ "stages": [
+  {
+   "stage_id": "17e-stage-1-governance",
+   "objective": "governance-only stage",
+   "state": "COMPLETE",
+   "depends_on": [],
+   "priority": 0
+  },
+  {
+   "stage_id": "17e-stage-2-async",
+   "objective": "async work under event bridge",
+   "state": "PENDING",
+   "depends_on": [],
+   "priority": 1
+  },
+  {
+   "stage_id": "17e-stage-3-fails",
+   "objective": "deliberately failing stage",
+   "state": "PENDING",
+   "depends_on": [],
+   "priority": 2
+  },
+  {
+   "stage_id": "17e-stage-4-final",
+   "objective": "final stage",
+   "state": "PENDING",
+   "depends_on": [],
+   "priority": 3
+  }
+ ],
+ "fallbacks": [],
+ "saved_at": "2026-09-15T17:50:29Z"
+}

tests/conftest.py

@@ -12,12 +12,49 @@ override via monkeypatch.setenv, which wins over this fixture.
 """
 
 import os
+import sys
+from pathlib import Path
 
 import pytest
 
+# W/R bridge code dir (daemon/waitstore live out-of-repo, own snapshot).
+# Env-overridable so runner users without $HOME access can point at an
+# installed copy — T15 legacy hardcode, resolved T16.
+WR_CODE_DIR = Path(os.environ.get(
+    "RUNE_WR_CODE", "/home/nick/.local/share/mimo-wait-resume"))
+if str(WR_CODE_DIR) not in sys.path:
+    sys.path.insert(0, str(WR_CODE_DIR))
+
+TESTS_DIR = Path(__file__).resolve().parent
+REPO_ROOT = TESTS_DIR.parent
+if str(REPO_ROOT) not in sys.path:
+    sys.path.insert(0, str(REPO_ROOT))
+
 
 @pytest.fixture(autouse=True)
 def _isolate_construction_store(tmp_path, monkeypatch):
     monkeypatch.setenv("RUNE_CONSTRUCTION_STORE",
                        str(tmp_path / ".rune" / "scaffold"))
     yield
+
+
+@pytest.fixture(autouse=True)
+def _provide_test_provider_context(monkeypatch):
+    """T16 §16F: drivers/probes resolve the model from the ambient
+    provider context (MIMO_MODEL) and fail loudly without one. In tests,
+    the HARNESS is the surrounding provider layer — so the default test
+    context provides one, unless a test explicitly removes it (the
+    PROVIDER_UNRESOLVED held-outs do exactly that)."""
+    monkeypatch.setenv("MIMO_MODEL", "test-provider/test-model")
+
+
+@pytest.fixture()
+def wr_daemon(tmp_path, monkeypatch):
+    """Real W/R daemon subprocess on an isolated WR_ROOT + ephemeral port.
+    Shared by test_event_wait.py (T15) and test_ci_event_wait.py (T16)."""
+    monkeypatch.setenv("RUNE_WR_ROOT", str(tmp_path))  # adapter binds here
+    monkeypatch.setenv("RUNE_WR_CODE", str(WR_CODE_DIR))
+    from tests.test_event_wait import Daemon
+    d = Daemon(tmp_path)
+    yield d
+    d.stop()

tests/test_ci_event_wait.py

@@ -0,0 +1,303 @@
+"""test_ci_event_wait.py — T16 §16C/§16D/§16E: event-driven CI
+qualification wait over the EXISTING W/R bridge.
+
+Same disposable harness as test_event_wait.py (real daemon subprocess on
+an isolated WR_ROOT + ephemeral port; Daemon fixture imported — 16W rule:
+no duplicated test helpers).
+
+Proofs:
+  16C event wake   — CI-completion event (provider=gitea,
+                     event_type=ci_completed, correlated by sha) resumes
+                     the waiting controller; the EVENT payload is NOT
+                     truth — the verdict is re-read from canonical Gitea
+                     authority (monkeypatched here; production = deploy_
+                     qualify.py).
+  16C event≠truth  — event claims success, authority says failure →
+                     deployment refused (payload never decides).
+  16D no floor     — event → resume latency is event-driven (sub-second),
+                     not a fixed sleep floor.
+  16D lost event   — event suppressed → watchdog expiry → ONE-SHOT
+                     authority query → truthful recovery (or truthful
+                     failure when authority has no verdict). Watchdog is
+                     the failure detector, never the scheduler.
+  16D re-arm       — watchdog expiry while CI runs are non-terminal in
+                     Gitea → re-arm, not failure; a later event resumes.
+  16E one primitive— builder and gitea waits share the SAME bridge
+                     machinery; identities select the matching events
+                     (a gitea event never resumes a builder wait and
+                     vice versa).
+"""
+
+from __future__ import annotations
+
+import importlib.util
+import json
+import sys
+import threading
+import time
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune import event_wait as ew  # noqa: E402
+from tests.test_event_wait import Daemon  # noqa: E402  (shared fixture)
+
+
+def _load_ci_wait_module():
+    spec = importlib.util.spec_from_file_location(
+        "ci_event_wait", str(REPO / "scripts" / "ci_event_wait.py"))
+    mod = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(mod)
+    return mod
+
+
+CI_SHA = "e" * 40
+
+
+def _post_when_registered(wr_daemon: Daemon, op: str, payload: dict,
+                          timeout: float = 10.0):
+    """Thread body: wait for the wait record to exist, then post the wake
+    event. The event POST is best-effort (tests assert on outcomes)."""
+
+    def go():
+        deadline = time.time() + timeout
+        while time.time() < deadline:
+            if ew.operation_wait(op) is not None:
+                break
+            time.sleep(0.05)
+        wr_daemon.post_event(payload)  # loud: a failed wake must fail the test
+
+    t = threading.Thread(target=go, daemon=True)
+    t.start()
+    return t
+
+
+def test_16c_event_wake_resumes_and_authority_decides(wr_daemon, monkeypatch):
+    mod = _load_ci_wait_module()
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "success")
+    monkeypatch.setattr(mod, "ci_alive", lambda: False)
+    payload = {"operation_id": f"ci-{CI_SHA}", "provider": "gitea",
+               "event_type": "ci_completed", "sha": CI_SHA, "exit_code": 0,
+               "completed_at": time.time()}
+    _post_when_registered(wr_daemon, f"ci-{CI_SHA}", payload)
+    t0 = time.monotonic()
+    rc = mod.run(sha=CI_SHA, deadline_secs=25, daemon_url=wr_daemon.url())
+    wall = time.monotonic() - t0
+    # rc==0 alone is NOT proof: watchdog recovery also returns 0. The wake
+    # must actually have driven the resume (fast), not the watchdog.
+    assert rc == 0
+    assert wall < 15, f"wake-resume took {wall:.1f}s — event never arrived"
+    rec = ew.operation_wait(f"ci-{CI_SHA}")
+    assert rec is not None and rec["provider"] == "gitea"
+    assert rec["wait_kind"] == "PROCESS_EVENT"
+
+
+def test_16c_event_payload_is_never_truth(wr_daemon, monkeypatch):
+    mod = _load_ci_wait_module()
+    # the event says success (exit_code 0); canonical authority says the
+    # CI is RED — the payload must never decide the verdict
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "failure")
+    monkeypatch.setattr(mod, "ci_alive", lambda: False)
+    payload = {"operation_id": f"ci-{CI_SHA}", "provider": "gitea",
+               "event_type": "ci_completed", "sha": CI_SHA, "exit_code": 0,
+               "completed_at": time.time()}
+    _post_when_registered(wr_daemon, f"ci-{CI_SHA}", payload)
+    rc = mod.run(sha=CI_SHA, deadline_secs=25, daemon_url=wr_daemon.url())
+    assert rc == 1  # refused, despite the wake saying 0
+
+
+def test_16d_event_resume_has_no_sleep_floor(wr_daemon, monkeypatch):
+    mod = _load_ci_wait_module()
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "success")
+    monkeypatch.setattr(mod, "ci_alive", lambda: False)
+    payload = {"operation_id": f"ci-{CI_SHA}", "provider": "gitea",
+               "event_type": "ci_completed", "sha": CI_SHA, "exit_code": 0,
+               "completed_at": time.time()}
+    _post_when_registered(wr_daemon, f"ci-{CI_SHA}", payload)
+    t0 = time.monotonic()
+    rc = mod.run(sha=CI_SHA, deadline_secs=25, daemon_url=wr_daemon.url())
+    wall = time.monotonic() - t0
+    # a would-be 30s polling floor is NOT waited: registration + event +
+    # verdict must be seconds, not the old sleep-30 loop quantum
+    assert rc == 0
+    assert wall < 15, f"event-driven wait took {wall:.1f}s — floor crept back"
+
+
+def test_16d_lost_event_watchdog_recovers_from_authority(
+        wr_daemon, monkeypatch):
+    mod = _load_ci_wait_module()
+    # NO event is ever posted (suppressed/lost); authority IS resolvable
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "success")
+    monkeypatch.setattr(mod, "ci_alive", lambda: False)
+    rc = mod.run(sha=CI_SHA, deadline_secs=4, daemon_url=wr_daemon.url())
+    assert rc == 0
+    # truthful recovery, not invented success: the record shows the
+    # watchdog (failure detector) path, and the verdict came from authority
+    rec = ew.operation_wait(f"ci-{CI_SHA}")
+    assert rec is not None
+
+
+def test_16d_lost_event_with_no_verdict_fails_truthfully(
+        wr_daemon, monkeypatch):
+    mod = _load_ci_wait_module()
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "pending")
+    monkeypatch.setattr(mod, "ci_alive", lambda: False)
+    rc = mod.run(sha=CI_SHA, deadline_secs=4, daemon_url=wr_daemon.url())
+    assert rc == 1  # deployment refused — timeout is failure detection
+
+
+def test_16d_ci_alive_rearm_then_event(wr_daemon, monkeypatch):
+    mod = _load_ci_wait_module()
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "success")
+    # first watchdog expiry (t=3s): CI runs still non-terminal → re-arm,
+    # not fail; the event arrives only AFTER that expiry (t≥4s)
+    alive = {"n": 0}
+
+    def flaky_alive():
+        alive["n"] += 1
+        return alive["n"] <= 1  # alive on first expiry, gone after event
+
+    monkeypatch.setattr(mod, "ci_alive", flaky_alive)
+
+    def post_late():
+        deadline = time.time() + 10
+        while time.time() < deadline:
+            if ew.operation_wait(f"ci-{CI_SHA}") is not None:
+                break
+            time.sleep(0.05)
+        time.sleep(4.0)  # strictly after the first watchdog expiry (3s)
+        wr_daemon.post_event({"operation_id": f"ci-{CI_SHA}",
+                              "provider": "gitea",
+                              "event_type": "ci_completed",
+                              "sha": CI_SHA, "exit_code": 0,
+                              "completed_at": time.time()})
+
+    threading.Thread(target=post_late, daemon=True).start()
+    rc = mod.run(sha=CI_SHA, deadline_secs=15, watchdog_secs=3,
+                 daemon_url=wr_daemon.url())
+    assert rc == 0
+    assert alive["n"] >= 1  # the re-arm actually happened
+
+
+def test_16e_one_bridge_two_adapters_identity_selects(wr_daemon):
+    """16E: builder and gitea waits share the same primitive; only the
+    identity fields differ, and identity selects the matching event."""
+    # builder-style wait
+    b = ew.register_process_wait("op-builder-1", watchdog_secs=60)
+    # gitea-style wait
+    g = ew.register_process_wait(f"ci-{CI_SHA}", provider="gitea",
+                                 event_type="ci_completed", sha=CI_SHA,
+                                 watchdog_secs=60)
+    assert b["provider"] == "builder" and g["provider"] == "gitea"
+    # a gitea CI event must NOT resume the builder wait...
+    wr_daemon.post_event({"operation_id": f"ci-{CI_SHA}", "provider": "gitea",
+                       "event_type": "ci_completed", "sha": CI_SHA,
+                       "exit_code": 0})
+    brec = ew.operation_wait("op-builder-1")
+    grec = ew.operation_wait(f"ci-{CI_SHA}")
+    assert brec["status"] == "WAITING_EXTERNAL"
+    assert grec["status"] in ("READY_TO_RESUME", "RESUME_UNVERIFIED",
+                              "COMPLETED")
+    # ...and a builder event must not resume a (fresh) gitea wait
+    g2 = ew.register_process_wait(f"ci-{'f' * 40}", provider="gitea",
+                                  event_type="ci_completed", sha="f" * 40,
+                                  watchdog_secs=60)
+    wr_daemon.post_event({"operation_id": "op-builder-1", "exit_code": 0})
+    assert ew.operation_wait(f"ci-{'f' * 40}")["status"] == "WAITING_EXTERNAL"
+    assert ew.operation_wait("op-builder-1")["status"] in (
+        "READY_TO_RESUME", "RESUME_UNVERIFIED", "COMPLETED")
+
+
+def test_16c_event_replay_before_registration(wr_daemon, monkeypatch):
+    """§16C live finding (serialized runner): the completion event lands
+    BEFORE the deploy job registers. The durable event log replays it —
+    the waiter skips the rendezvous and goes straight to authority."""
+    mod = _load_ci_wait_module()
+    monkeypatch.setattr(mod, "authority_verdict", lambda: "success")
+    monkeypatch.setattr(mod, "ci_alive", lambda: False)
+    wr_daemon.post_event({"operation_id": f"ci-{CI_SHA}",
+                       "provider": "gitea", "event_type": "ci_completed",
+                       "sha": CI_SHA, "exit_code": 0,
+                       "completed_at": time.time()})
+    t0 = time.monotonic()
+    rc = mod.run(sha=CI_SHA, deadline_secs=60, watchdog_secs=300,
+                 daemon_url=wr_daemon.url())
+    wall = time.monotonic() - t0
+    assert rc == 0
+    assert wall < 10, f"replayed wake took {wall:.1f}s — replay path dead"
+    assert ew.find_matching_event("gitea", "ci_completed",
+                                  f"ci-{CI_SHA}") is not None
+
+
+def test_16c_replay_does_not_match_other_identities(wr_daemon):
+    wr_daemon.post_event({"operation_id": f"ci-{CI_SHA}",
+                       "provider": "gitea", "event_type": "ci_completed",
+                       "sha": CI_SHA, "exit_code": 0})
+    assert ew.find_matching_event("gitea", "ci_completed",
+                                  "ci-other") is None
+    assert ew.find_matching_event("builder", "builder_exit",
+                                  f"ci-{CI_SHA}") is None
+
+
+def test_17c_fifo_mode_is_cross_user(wr_daemon):
+    """T17 17C ROOT-CAUSE REGRESSION GUARD: the T16 chmod(0o666) was lost
+    in a rebase; FIFOs landed 644, the daemon (different user) got EACCES
+    on the resume write, and every production waiter slept to its
+    deadline. Mode must be 666 regardless of umask/creating user."""
+    import os as _os
+    import stat as _stat
+    fd = ew.open_fifo_reader("ci-mode-guard-op")
+    try:
+        mode = _os.stat(ew.fifo_path("ci-mode-guard-op")).st_mode
+        assert mode & 0o666 == 0o666, (
+            f"fifo mode {oct(mode & 0o777)} — resume writer from another "
+            f"user will get EACCES (17C root-cause regression)")
+    finally:
+        _os.close(fd)
+
+
+def test_17c_cross_user_resume_delivery(wr_daemon):
+    """Full cross-user delivery: the WAITER side (fifo + registration)
+    runs as the invoking user; the resume WRITE runs through the real
+    daemon (as its own user in production). After the mode fix the
+    daemon's writer must succeed (rc=0, wait COMPLETED), not rc=1."""
+    import os as _os
+    op = f"ci-crossuser-{int(time.time())}"
+    fd = ew.open_fifo_reader(op)
+    rec = ew.register_process_wait(op, provider="gitea",
+                                   event_type="ci_completed",
+                                   sha="crossuser", watchdog_secs=60)
+    try:
+        import threading
+        done = {}
+
+        def waiter():
+            done["r"] = ew.await_resume_event(op, watchdog_secs=15,
+                                              fifo_fd=fd)
+
+        t = threading.Thread(target=waiter, daemon=True)
+        t.start()
+        time.sleep(0.3)
+        wr_daemon.post_event({"operation_id": op, "provider": "gitea",
+                              "event_type": "ci_completed",
+                              "sha": "crossuser", "exit_code": 0,
+                              "completed_at": time.time()})
+        t.join(timeout=20)
+        assert done.get("r", {}).get("kind") == "event", (
+            f"resume delivery failed again: {done} — 17C root-cause "
+            f"regression")
+        # the durable record must show the writer exited rc=0
+        deadline = time.time() + 10
+        while time.time() < deadline:
+            r = ew.operation_wait(op) or {}
+            if r.get("status") == "COMPLETED":
+                break
+            time.sleep(0.2)
+        assert r.get("status") == "COMPLETED", (
+            f"wait status {r.get('status')} — resume writer did not "
+            f"deliver cleanly (expect COMPLETED, not RESUME_UNVERIFIED)")
+    finally:
+        _os.close(fd)

tests/test_construction_interface.py

@@ -549,3 +549,39 @@ def test_capabilities_cli_roundtrip(tmp_path):
     assert out["success"]
     assert out["state"]["schema_version"] == ci.INTERFACE_SCHEMA_VERSION
     assert "capabilities" in out["state"]["operations"]
+
+
+def test_15u_discover_active_contract_migrated_to_interface(tmp_path):
+    """§15U simplification: resume discovery lives in the INTERFACE (one
+    canonical home), not in session-driver scripts. The drivers' helpers
+    are shims over this."""
+    from rune import construction_interface as ci
+    store = tmp_path / "scaffold"
+    iface = ci.ConstructionInterface(builder="probe",
+                                     project_store=str(store),
+                                     project_id="t15u")
+    iface.scaffold.init(purpose="15u probe")
+    empty = ci.ConstructionInterface.discover_active_contract(str(store))
+    assert not empty["success"], "empty store must not invent a contract"
+    doc = iface.scaffold.load()
+    from rune import structural_store as ss
+    ss.add_entity(doc, id="contract:spec-15u", kind="contract",
+                  identities={"objective": "spec"}, evidence_refs=["e"])
+    ss.add_entity(doc, id="capability:planned", kind="capability",
+                  maturity="DECLARED", evidence_refs=["e"])
+    ss.add_relationship(doc, type="implements",
+                        from_id="capability:planned",
+                        to_id="contract:spec-15u", evidence_refs=["e"])
+    iface.scaffold.commit(doc, what="seed 15u spec")
+    sub = iface.submit_need(
+        "build the 15u module",
+        constraints={"environment": {"os": "linux"},
+                     "contracts": ["contract:spec-15u"]},
+        evidence_refs=["e"], approval_context="15u")
+    assert sub["success"]
+    found = ci.ConstructionInterface.discover_active_contract(str(store))
+    assert found["success"]
+    assert found["ids"]["contract_id"] == sub["ids"]["contract_id"]
+    # capabilities() advertises the new op
+    assert "discover_active_contract" in \
+        iface.capabilities()["state"]["operations"]

tests/test_drift_resume.py

@@ -0,0 +1,82 @@
+"""test_drift_resume.py — T16 §16P: drift-aware resume held-outs,
+driven by the REAL 16O fixture."""
+
+from __future__ import annotations
+
+import json
+import subprocess
+import sys
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune import drift_resume as dr  # noqa: E402
+
+FIXTURE = Path("/mnt/workdrive/t16-reentry")
+REPO_DIR = FIXTURE / "repo"
+STORE = FIXTURE / "scaffold" / "current.json"
+
+
+def test_16p_drift_classes_deterministic():
+    ok = {"actual": "", "checked_by": "pytest", "matches": True}
+    bad = {"actual": "2 tests fail", "checked_by": "pytest", "matches": False}
+    unverif = {"actual": None, "checked_by": "pytest", "matches": None}
+    assert dr.classify_drift({"claim": "green"}, ok)["class"] == "CONFIRMED"
+    assert dr.classify_drift({"claim": "green"}, bad)["class"] == "MATERIAL"
+    assert dr.classify_drift({"claim": "x"}, unverif)["class"] == "UNKNOWN"
+    locked = dict({"claim": "layout"}, user_locked=True)
+    assert dr.classify_drift(locked, bad)["class"] == "CONFLICTING"
+
+
+def test_16p_reality_wins_user_locked_needs_user():
+    r = dr.reconcile(
+        [{"claim": "layout locked as X", "user_locked": True}],
+        lambda c: {"actual": "layout is Y", "checked_by": "ls",
+                   "matches": False})
+    assert r["resume"] is False          # blocked, not overridden
+    assert r["blocking"][0]["class"] == "CONFLICTING"
+
+
+def test_16p_real_fixture_vat_regression_blocks_resume():
+    """The REAL fixture: scaffold claims stage-1 verified green; reality
+    has the VAT regression. The reconciliation must block."""
+    doc = json.loads(STORE.read_text())
+    stage1 = next(e for e in doc["entities"]
+                  if e["id"] == "contract:stage-1-monthly-totals")
+
+    def run_fixture_tests(claim):
+        proc = subprocess.run(
+            [sys.executable, "-m", "pytest", "tests/", "-q"],
+            cwd=REPO_DIR, capture_output=True, text=True,
+            env={"PATH": "/usr/bin:/bin:/usr/local/bin",
+                 "PYTHONPATH": "src"})
+        passed = proc.returncode == 0
+        return {"actual": proc.stdout.strip().splitlines()[-1]
+                if proc.stdout else proc.stderr[:120],
+                "checked_by": "pytest tests/", "matches": passed}
+
+    claims = [{"claim": "stage-1 verified green", "kind": "verification",
+               "stale_ok": False}]
+    r = dr.reconcile(claims, run_fixture_tests)
+    assert r["resume"] is False
+    assert r["blocking"][0]["class"] == "MATERIAL"
+    assert "failed" in r["blocking"][0]["evidence"]
+    assert stage1["phase"] == "closed"  # the stale claim we caught
+
+
+def test_16p_clean_repo_resumes():
+    """Same observer on the checkpoint commit: no drift, resume allowed."""
+    def check_at_checkpoint(claim):
+        proc = subprocess.run(
+            ["git", "show", "39bd917:src/report.py"], cwd=REPO_DIR,
+            capture_output=True, text=True)
+        vat_free = "1.19" not in proc.stdout
+        return {"actual": "" if vat_free else "VAT present at checkpoint",
+                "checked_by": "git show", "matches": vat_free}
+
+    r = dr.reconcile([{"claim": "stage-1 green at checkpoint"}],
+                     check_at_checkpoint)
+    assert r["resume"] is True and r["blocking"] == []

tests/test_event_wait.py

@@ -0,0 +1,260 @@
+"""test_event_wait.py — T15 §15D/§15E: event-driven wait as canonical
+default + event semantics adversarials.
+
+Runs the REAL W/R daemon as a subprocess against an isolated WR_ROOT
+(the same disposable harness pattern as the bridge's own v0.2/v0.3
+invariant tests), then drives the repo-side adapter (rune.event_wait)
+through it. No real MiMo sessions; the "builder" is a cheap fake command.
+
+Proofs:
+  15D latency      — controller resumes on the EVENT; continuation
+                     latency recorded; no arbitrary wait floor exists
+                     (a deliberately long would-be floor is NOT waited).
+  15D failure      — non-zero builder exit → verdict failure, resume
+                     still fires (timeout is not success; events carry
+                     truth).
+  15D watchdog     — lost event → deadline expires → one-shot liveness
+                     inspection → PROCESS_LOST classification. Watchdog
+                     is the failure detector, never the scheduler.
+  15E duplicate    — duplicate completion event cannot double-resume.
+  15E stale        — event from a prior operation matches nothing.
+  15E wrong corr   — unknown operation_id matches nothing.
+  15E malformed    — event without exit_code leaves wait WAITING.
+  15E restart      — controller restart after event persisted: durable
+                     waitstore wins; resume_or_recover is idempotent.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import signal
+import subprocess
+import sys
+import tempfile
+import time
+import urllib.request
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+WR = Path("/home/nick/.local/share/mimo-wait-resume")
+sys.path.insert(0, str(WR))
+
+from rune import event_wait as ew  # noqa: E402
+
+
+class Daemon:
+    """Real daemon subprocess on an isolated WR_ROOT + ephemeral port."""
+
+    def __init__(self, tmp_root: Path, port: int | None = None):
+        # Ephemeral port (not a fixed one): the CI runner executes the
+        # FULL suite in one process, and a fixed port collides with
+        # TIME_WAIT sockets from earlier tests in the same file or with
+        # parallel jobs on the same runner. 15Z CI failure: the latency
+        # test's daemon failed to bind 8793 (stale socket), the notifier's
+        # event POSTs went nowhere, and the watchdog correctly reported
+        # PROCESS_LOST — the test harness was at fault, not the bridge.
+        if port is None:
+            import socket as _socket
+            with _socket.socket() as _s:
+                _s.bind(("127.0.0.1", 0))
+                port = _s.getsockname()[1]
+        self.root, self.port = tmp_root, port
+        env = dict(os.environ,
+                   WR_ROOT=str(tmp_root),
+                   WR_DRIVE_SECS="1",
+                   WR_RECONCILE_SECS="2",
+                   WR_RECONCILE_TICKS="1",
+                   WR_WATCH_SECS="30",
+                   WR_TERM_GRACE_SECS="1",
+                   RUNE_WR_ROOT=str(tmp_root))
+        self.proc = subprocess.Popen(
+            [sys.executable, str(WR / "daemon.py"), str(port)],
+            stdout=subprocess.PIPE, stderr=subprocess.STDOUT,
+            text=True, env=env)
+        deadline = time.time() + 15
+        while time.time() < deadline:
+            try:
+                with urllib.request.urlopen(
+                        f"http://127.0.0.1:{port}/health", timeout=2) as r:
+                    if r.status == 200:
+                        return
+            except Exception:
+                time.sleep(0.2)
+        raise RuntimeError("daemon did not become healthy")
+
+    def url(self) -> str:
+        return f"http://127.0.0.1:{self.port}"
+
+    def stop(self):
+        self.proc.send_signal(signal.SIGTERM)
+        try:
+            self.proc.wait(timeout=5)
+        except subprocess.TimeoutExpired:
+            self.proc.kill()
+
+    def post_event(self, payload: dict) -> dict:
+        req = urllib.request.Request(
+            self.url() + "/process-event",
+            data=json.dumps(payload).encode(),
+            headers={"Content-Type": "application/json"})
+        with urllib.request.urlopen(req, timeout=5) as r:
+            return json.loads(r.read())
+
+
+# T16 §16E: the daemon fixture moved to tests/conftest.py (wr_daemon) so
+# test_ci_event_wait.py shares ONE harness — no duplicated test helpers.
+@pytest.fixture()
+def daemon(wr_daemon):
+    return wr_daemon
+
+
+# real fake-builder: sleeps then exits — cheap but genuinely a process
+FAKE_BUILDER = ["bash", "-c", "sleep 2; exit 0"]
+FAKE_BUILDER_FAIL = ["bash", "-c", "sleep 1; exit 3"]
+
+
+def test_15d_latency_event_driven_no_floor(daemon):
+    """Continuation happens FROM the event: controller resumes ~immediately
+    after the 2s builder exits — proving no 60s+ polling floor exists."""
+    op = "op-latency-proof"
+    rec = ew.register_process_wait(
+        op, contract_id="contract:t15", project="t15-fixture",
+        objective="latency proof", watchdog_secs=120)
+    assert rec["wait_kind"] == "PROCESS_EVENT"
+    assert rec["resume_cmd"], "per-wait resume_cmd required"
+
+    builder_started = time.time()
+    notifier = ew.run_builder_under_events(FAKE_BUILDER, op,
+                                           daemon_url=daemon.url())
+    result = ew.await_resume_event(op, watchdog_secs=120,
+                                   pgid=notifier.pid)
+    resumed_at = time.time()
+    assert result["kind"] == "event", f"expected event resume, got {result}"
+    wait = result["wait"]
+    assert wait["verdict"] == "success"
+    evidence = wait.get("process_event_evidence", {})
+    child_exit = evidence and float(evidence.get("exit_code") == 0)
+    # The event carries the actual exit evidence
+    assert evidence.get("exit_code") == 0
+    builder_elapsed = resumed_at - builder_started
+    # builder itself takes ~2s; if a 60s polling floor existed we'd exceed it
+    assert builder_elapsed < 30, \
+        f"continuation took {builder_elapsed:.1f}s — a wait floor exists"
+    # and it cannot be faster than the builder itself (event came from
+    # the real process exit, not a timer)
+    assert builder_elapsed >= 1.5
+    assert wait["status"] in ("RESUMING", "COMPLETED")
+
+
+def test_15d_failure_event_carries_truth(daemon):
+    op = "op-failure-proof"
+    ew.register_process_wait(op, project="t15-fixture",
+                             objective="failure proof", watchdog_secs=60)
+    notifier = ew.run_builder_under_events(FAKE_BUILDER_FAIL, op,
+                                           daemon_url=daemon.url())
+    result = ew.await_resume_event(op, watchdog_secs=60, pgid=notifier.pid)
+    assert result["kind"] == "event"
+    assert result["wait"]["verdict"] == "failure"
+    assert result["wait"]["process_event_evidence"]["exit_code"] == 3
+
+
+def test_15d_watchdog_detects_lost_event(daemon, tmp_path):
+    """Lost event: deadline expires → one-shot liveness inspection →
+    PROCESS_LOST. The watchdog never schedules; it detects failure of
+    continuation."""
+    op = "op-lost-event"
+    rec = ew.register_process_wait(op, project="t15-fixture",
+                                   objective="lost event", watchdog_secs=3)
+    # launch a long-running builder whose event will NEVER arrive
+    notifier = ew.run_builder_under_events(
+        ["bash", "-c", "sleep 120"], op, daemon_url=daemon.url())
+    result = ew.await_resume_event(op, watchdog_secs=3, pgid=notifier.pid)
+    assert result["kind"] == "watchdog"
+    assert result["alive"] is True  # process still alive → NOT lost, re-arm
+    # now genuinely lose it: kill the group, next watchdog sees GONE
+    os.killpg(notifier.pid, signal.SIGKILL)
+    result2 = ew.await_resume_event(op, watchdog_secs=3, pgid=notifier.pid)
+    assert result2["kind"] == "watchdog"
+    assert result2["alive"] is False  # PROCESS_LOST → recovery path
+    # recovery: process died without completion — the wait is still
+    # WAITING_EXTERNAL (canonical state: nothing completed)
+    wait = ew.operation_wait(op)
+    assert wait["status"] == "WAITING_EXTERNAL"
+
+
+def test_15e_duplicate_event_cannot_double_resume(daemon):
+    op = "op-dup-event"
+    ew.register_process_wait(op, project="t15-fixture", watchdog_secs=60)
+    first = daemon.post_event({"operation_id": op, "exit_code": 0})
+    assert first["matched"] >= 1
+    wait_for(lambda: (ew.operation_wait(op)["status"] != "READY_TO_RESUME"))
+    second = daemon.post_event({"operation_id": op, "exit_code": 0})
+    # second event matches nothing new (wait no longer WAITING_EXTERNAL)
+    # and must not re-dispatch a resumed wait
+    rec = ew.operation_wait(op)
+    if second["matched"] > 0:
+        # if it dispatched again, it must NOT have reset resume state
+        assert rec.get("resume_attempts", 0) >= 1
+    assert rec["status"] != "WAITING_EXTERNAL"
+    # verdict unchanged by the duplicate
+    assert rec.get("verdict") in ("success", None)
+
+
+def test_15e_stale_event_from_prior_operation_matches_nothing(daemon):
+    op_old, op_new = "op-old", "op-new"
+    ew.register_process_wait(op_old, project="t15-fixture", watchdog_secs=60)
+    ew.register_process_wait(op_new, project="t15-fixture", watchdog_secs=60)
+    # satisfy + consume the old operation's event
+    daemon.post_event({"operation_id": op_old, "exit_code": 0})
+    wait_for(lambda: ew.operation_wait(op_old)["status"] != "READY_TO_RESUME")
+    # a STALE repeat of the old event must not touch the new operation
+    daemon.post_event({"operation_id": op_old, "exit_code": 0})
+    new_wait = ew.operation_wait(op_new)
+    assert new_wait["status"] == "WAITING_EXTERNAL"
+    assert new_wait.get("verdict") is None
+
+
+def test_15e_wrong_correlation_matches_nothing(daemon):
+    op = "op-wrong-corr"
+    ew.register_process_wait(op, project="t15-fixture", watchdog_secs=60)
+    resp = daemon.post_event({"operation_id": "completely-unrelated",
+                              "exit_code": 0})
+    assert resp["matched"] == 0
+    assert ew.operation_wait(op)["status"] == "WAITING_EXTERNAL"
+
+
+def test_15e_malformed_event_without_exit_code_stays_waiting(daemon):
+    op = "op-malformed"
+    ew.register_process_wait(op, project="t15-fixture", watchdog_secs=60)
+    resp = daemon.post_event({"operation_id": op})  # no exit_code
+    assert ew.operation_wait(op)["status"] == "WAITING_EXTERNAL"
+
+
+def test_15e_restart_after_event_persisted_waitstore_wins(daemon):
+    """Case 10: the event was persisted and the wait advanced, but the
+    controller 'crashed' before handling it. On restart the durable
+    waitstore wins; resume_or_recover is idempotent (no duplicate
+    transition, no duplicate resume)."""
+    op = "op-restart"
+    ew.register_process_wait(op, project="t15-fixture", watchdog_secs=60)
+    daemon.post_event({"operation_id": op, "exit_code": 0})
+    wait_for(lambda: ew.operation_wait(op)["status"] != "READY_TO_RESUME")
+    r1 = ew.resume_or_recover(op)
+    r2 = ew.resume_or_recover(op)
+    assert r1["kind"] == "event" and r1.get("recovered") is True
+    assert r2 == r1  # idempotent: canonical state, not transient token
+    assert r1["wait"]["verdict"] == "success"
+
+
+def wait_for(fn, timeout=30, interval=0.3):
+    deadline = time.time() + timeout
+    while time.time() < deadline:
+        r = fn()
+        if r:
+            return r
+        time.sleep(interval)
+    return fn()

tests/test_native_activation.py

@@ -19,6 +19,14 @@ def native_operator(tmp_path, monkeypatch):
     stubbed so the test exercises only the construction hook."""
     store = tmp_path / ".rune" / "scaffold"
     monkeypatch.setenv("RUNE_CONSTRUCTION_STORE", str(store))
+    # §15X: hook tests exercise the construction hook, not the provider —
+    # stub the downstream model execution (send_task otherwise makes REAL
+    # model calls, burning quota and coupling unit tests to provider health)
+    from rune.tool_protocol import OperatorResponse
+    monkeypatch.setattr(
+        "rune.rune_operator.RuneOperator._execute_on_model",
+        lambda self, *a, **k: OperatorResponse(success=True,
+                                               user_text="stubbed"))
     from rune import scaffold as sc
     sc.Scaffold(store, "rune").init(purpose="rune native test")
     import rune.rune_operator as ro

tests/test_native_default.py

@@ -0,0 +1,251 @@
+"""test_native_default.py — T15 §15J/§15K: Scaffolding as the NATIVE
+DEFAULT construction route + bypass prevention.
+
+15J: ordinary send_task input (the real TUI submission path per the
+12A.C trace: TUI HistoryInput → submit → RuneOperator.send_task →
+construction hook) automatically produces canonical construction state:
+  A. PROJECT_BUILD  — durable Project Scaffold, no /build command
+  B. EXISTING       — attaches Rune authority, no fake project
+  C. LOCAL_CHANGE   — lightweight, still governed
+  D. NOT_CONSTRUCTION — no construction state
+  E. UNKNOWN        — probes, no invention
+  F. manual override through the SAME path (/scaffold-mode marker)
+
+15K: a native client construction command (/build start) cannot mint
+construction outside the authority — NOT_CONSTRUCTION is refused;
+BuildManager state is created only AFTER canonical recognition.
+"""
+
+import json
+import os
+import sys
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune import scaffold as sc  # noqa: E402
+from rune import structural_store as ss  # noqa: E402
+
+
+@pytest.fixture()
+def op(tmp_path, monkeypatch):
+    monkeypatch.setenv("RUNE_CONSTRUCTION_STORE",
+                       str(tmp_path / ".rune" / "scaffold"))
+    from rune.rune_operator import RuneOperator
+    # §15X process repair: construction-hook tests exercise the hook, not
+    # the provider. send_task invokes a REAL model downstream of the hook —
+    # unstubbed, these unit tests burn provider quota and starve other
+    # suites' model calls (observed: stage-telemetry GATE_FAILED after
+    # quota exhaustion). The hook runs BEFORE model selection, so a stub
+    # changes nothing these tests assert.
+    from rune.tool_protocol import OperatorResponse
+    monkeypatch.setattr(
+        "rune.rune_operator.RuneOperator._execute_on_model",
+        lambda self, *a, **k: OperatorResponse(success=True,
+                                               user_text="stubbed"))
+    return RuneOperator(str(tmp_path / "operator.json")), tmp_path
+
+
+def _last_activation(store: Path):
+    doc = json.loads((store / ".rune" / "scaffold" / "current.json").read_text())
+    recs = [o for o in doc.get("observations", [])
+            if o.get("kind") == "construction_activated"]
+    return recs[-1] if recs else None
+
+
+# ── 15J ─────────────────────────────────────────────────────────────────────
+
+def test_15j_native_project_build_without_any_command(op):
+    op, tmp = op
+    op.send_task("Build me a standalone practice scheduler app")
+    rec = _last_activation(tmp)
+    assert rec is not None, "native input must produce construction state"
+    assert rec["mode"] == "PROJECT_BUILD"
+    assert rec["activated"] is True
+    # durable provisional Project Scaffold exists
+    projects = (tmp / ".rune" / "scaffold" / "projects")
+    assert any(projects.glob("*/current.json")), "identity before repository"
+
+
+def test_15j_native_existing_project_change_attaches_rune(op):
+    op, tmp = op
+    op.send_task("Add CSV export to the rune project")
+    rec = _last_activation(tmp)
+    assert rec["mode"] == "EXISTING_PROJECT_CHANGE"
+    assert rec["activated"] is True
+
+
+def test_15j_native_local_change_is_light(op):
+    op, tmp = op
+    op.send_task("Fix the typo in the README")
+    rec = _last_activation(tmp)
+    assert rec["mode"] == "LOCAL_CHANGE"
+    assert rec["activated"] is True
+
+
+def test_15j_native_not_construction_touches_nothing(op):
+    op, tmp = op
+    op.send_task("Explain modal interchange theory")
+    rec = _last_activation(tmp)
+    assert rec["mode"] == "NOT_CONSTRUCTION"
+    assert rec["activated"] is False
+    # no project invention
+    projects = (tmp / ".rune" / "scaffold" / "projects")
+    assert not projects.exists() or not any(projects.glob("*/current.json"))
+
+
+def test_15j_native_unknown_probes_without_invention(op):
+    op, tmp = op
+    op.send_task("Build something useful for this")
+    rec = _last_activation(tmp)
+    assert rec["mode"] == "UNKNOWN"
+    assert rec["activated"] is False
+    assert not (tmp / ".rune" / "scaffold" / "projects").exists() or \
+        not any((tmp / ".rune" / "scaffold" / "projects").glob("*/current.json"))
+
+
+def test_15j_manual_override_same_path(op):
+    """Case F: /scaffold-mode marker feeds explicit_override through the
+    SAME classify_mode call — provenance carries 'user'."""
+    op, tmp = op
+    op.send_task("/scaffold-mode NOT_CONSTRUCTION build me a scheduler app")
+    rec = _last_activation(tmp)
+    assert rec["mode"] == "NOT_CONSTRUCTION", \
+        "override must win through the SAME path"
+    assert "user" in " ".join(rec.get("classification_evidence", [])), \
+        "override provenance must be recorded"
+    # and a non-overridden version of the same text classifies normally
+    op.send_task("Build me a standalone practice scheduler app")
+    rec2 = _last_activation(tmp)
+    assert rec2["mode"] == "PROJECT_BUILD"
+
+
+# ── 15K ─────────────────────────────────────────────────────────────────────
+
+def test_15k_build_manager_has_no_parallel_construction_state(tmp_path,
+                                                              monkeypatch):
+    """§15K inventory: BuildManager (the /build vehicle) creates its own
+    phase state but must NOT create project scaffolds/contracts. The
+    canonical construction record comes only from the send_task hook.
+
+    BuildManager.start_build shells out to `git worktree add` against a
+    HARDCODED production repo path (/mnt/workdrive/rune) — that is real
+    VCS mutation and has no place in a unit test (it fails on CI with
+    dubious-ownership, and would pollute the real repo locally). This
+    test therefore inspects the manager's state machine WITHOUT invoking
+    the worktree-creating path: construction-state separation is a
+    property of where BuildManager writes, not of the git operation."""
+    from rune import build_mode
+    from rune.build_mode import BuildManager
+    monkeypatch.chdir(tmp_path)  # NOT bare os.chdir — must not leak
+    monkeypatch.setattr(build_mode, "BUILDS_DIR",
+                        str(tmp_path / ".rune" / "work" / "builds"))
+    monkeypatch.setattr(build_mode, "WORKTREE_DIR",
+                        str(tmp_path / ".rune" / "worktrees"))
+    bm = BuildManager()
+    # start_build's worktree step is a real VCS mutation — stub it; the
+    # property under test (no canonical construction state) is orthogonal
+    monkeypatch.setattr(
+        BuildManager, "create_worktree",
+        lambda self: setattr(self.state, "worktree_path", "stubbed") or "stubbed")
+    build_id = bm.start_build("build a thing")
+    # BuildManager's own state exists...
+    assert bm.state.description == "build a thing"
+    # ...but it created NO canonical construction authority
+    assert not (tmp_path / "store" / "scaffold").exists() or \
+        not any((tmp_path / "store" / "scaffold").rglob("current.json"))
+
+
+def test_15k_tui_build_start_refuses_not_construction(monkeypatch, tmp_path):
+    """/build start with a non-construction description must be refused
+    BEFORE BuildManager mints build state — no silent bypass."""
+    monkeypatch.setenv("RUNE_CONSTRUCTION_STORE",
+                       str(tmp_path / ".rune" / "scaffold"))
+    from rune.rune_operator import RuneOperator
+    op = RuneOperator(str(tmp_path / "operator.json"))
+    desc = "explain modal interchange theory"
+    # the same classification the hardened /build start handler performs
+    from rune.construction_mode import classify_mode
+    d = classify_mode(desc, target_project=None, project_exists=None,
+                      affected_surface=[])
+    assert d.mode == "NOT_CONSTRUCTION"
+    # the handler's contract: refuse (never reach BuildManager.start_build)
+    # — proven here at the decision level plus the hook's canonical record
+    op.send_task(f"/build start: {desc}")
+    # via the hook path, the canonical store records NOT_CONSTRUCTION and
+    # no build artifact exists
+    doc = json.loads((tmp_path / ".rune" / "scaffold" / "current.json").read_text())
+    recs = [o for o in doc.get("observations", [])
+            if o.get("kind") == "construction_activated"]
+    assert recs and recs[-1]["mode"] in ("NOT_CONSTRUCTION", "UNKNOWN"), \
+        "prefixed build command must not mint construction for a chat intent"
+    assert recs[-1]["activated"] is False
+    assert not (tmp_path / ".rune" / "work" / "builds").exists() or \
+        not list((tmp_path / ".rune" / "work" / "builds").iterdir())
+
+
+def test_15l_research_requests_never_become_project_builds():
+    """15L negative: research/investigation intents must not PROJECT_BUILD
+    merely because they contain 'build'/'construct' words. Ownership:
+    research competence = Research Core; continuity = future Substrate.
+    The classifier may probe (UNKNOWN) but must never invent a project."""
+    from rune.construction_mode import classify_mode
+    research_intents = [
+        "investigate why the build fails",
+        "build understanding of the migration path",
+        "construct an explanation of modal interchange",
+        "research the best approach for CSV export",
+    ]
+    for t in research_intents:
+        d = classify_mode(t, target_project=None, project_exists=None,
+                          affected_surface=[])
+        assert d.mode != "PROJECT_BUILD", \
+            f"research intent '{t}' must not PROJECT_BUILD"
+        assert d.mode in ("UNKNOWN", "NOT_CONSTRUCTION")
+
+
+def test_15n_fresh_builder_discovers_observation_schema():
+    """15N: a fresh builder with NO prior knowledge queries capabilities()
+    first and learns the observation schema deterministically — the T14
+    dict-vs-list + missing-entity mistakes become discoverable, not
+    trial-and-error."""
+    import tempfile
+    from rune import construction_interface as ci
+    tmp = Path(tempfile.mkdtemp())
+    iface = ci.ConstructionInterface(builder="fresh-builder",
+                                     project_store=str(tmp / "s"),
+                                     project_id="t")
+    iface.scaffold.init(purpose="15N probe")
+    caps = iface.capabilities()["state"]
+    obs_schema = caps["argument_schemas"]["report_observation"]["observations"]
+    assert obs_schema["type"] == "list[dict]"
+    assert "entity" in obs_schema["item_required_fields"]
+    # guided by the schema, the builder's FIRST report is well-formed:
+    # seed a spec contract so the fit gate resolves (same proven pattern
+    # as the T14 fixture: the Need references the governed requirement)
+    from rune import structural_store as ss
+    doc = iface.scaffold.load()
+    ss.add_entity(doc, id="contract:spec-15n", kind="contract",
+                  identities={"objective": "spec"}, evidence_refs=["e"])
+    ss.add_entity(doc, id="capability:planned-15n", kind="capability",
+                  maturity="DECLARED", evidence_refs=["e"])
+    ss.add_relationship(doc, type="implements",
+                        from_id="capability:planned-15n",
+                        to_id="contract:spec-15n", evidence_refs=["e"])
+    iface.scaffold.commit(doc, what="seed 15n spec")
+    sub = iface.submit_need(
+        "build the 15n module",
+        constraints={"environment": {"os": "linux"},
+                     "contracts": ["contract:spec-15n"]},
+        evidence_refs=["e"], approval_context="15n")
+    assert sub["success"], sub.get("errors")
+    cid = sub["ids"]["contract_id"]
+    r = iface.report_observation(cid, [
+        {"entity": "m.js", "detail": "created"}])
+    assert r["success"], r.get("errors")
+    # and a schema-violating first attempt is REJECTED, not silently kept
+    bad = iface.report_observation(cid, {"entity": "x", "detail": "dict"})
+    assert not bad["success"]

tests/test_programme_continuation.py

@@ -0,0 +1,150 @@
+"""test_programme_continuation.py — T15 §15M: 'task done' ≠ 'programme
+done'.
+
+The required proof: a 4-stage programme where each completed stage
+AUTOMATICALLY starts the next — no user re-prompt, no voluntary stop.
+Plus the stop-classification matrix (every voluntary stop must carry an
+honest verdict) and the BLOCKED→fallback path.
+"""
+
+import json
+import sys
+from pathlib import Path
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune.programme_continuation import (  # noqa: E402
+    APPROVAL_REQUIRED, BLOCKED, COMPLETE, ELIGIBLE, FAILED, PENDING,
+    WAITING_EVENT, Programme, ProgrammeStage, run_programme,
+)
+
+
+def _four_stage_programme():
+    return Programme(
+        programme_id="prog-15m-proof",
+        objective="prove automatic continuation",
+        stages=[
+            ProgrammeStage("s1", "stage one", priority=1),
+            ProgrammeStage("s2", "stage two", depends_on=["s1"], priority=2),
+            ProgrammeStage("s3", "stage three waits for builder",
+                           depends_on=["s2"], priority=3),
+            ProgrammeStage("s4", "stage four", depends_on=["s3"], priority=4),
+        ])
+
+
+def test_15m_stage_completion_starts_next_stage_automatically():
+    """THE required proof: completing stage 1 MUST begin stage 2 without
+    a user re-prompt — the canary returns PROGRAMME_OPEN with a concrete
+    next action, never a stop."""
+    prog = _four_stage_programme()
+    v = prog.continuation_canary()
+    assert v["verdict"] == "PROGRAMME_OPEN"
+    assert v["next_action"] == ("start", "s1")
+    # stage 1 finishes...
+    prog.stages[0].state = COMPLETE
+    v2 = prog.continuation_canary()
+    assert v2["verdict"] == "PROGRAMME_OPEN"
+    assert v2["next_action"] == ("start", "s2"), \
+        "stage 2 must start immediately — no manual continue"
+    # stage 2 finishes → stage 3...
+    prog.stages[1].state = COMPLETE
+    v3 = prog.continuation_canary()
+    assert v3["next_action"] == ("start", "s3")
+    # stage 3 finishes → stage 4 → programme complete
+    prog.stages[2].state = COMPLETE
+    prog.stages[3].state = COMPLETE
+    v4 = prog.continuation_canary()
+    assert v4["verdict"] == "PROGRAMME_COMPLETE"
+
+
+def test_15m_event_wait_is_open_not_stop():
+    prog = _four_stage_programme()
+    prog.stages[0].state = COMPLETE
+    prog.stages[1].state = COMPLETE
+    prog.stages[2].state = WAITING_EVENT
+    v = prog.continuation_canary()
+    assert v["verdict"] == "PROGRAMME_OPEN_WAITING"
+    assert v["next_action"] == ("await_event", "s3")
+    assert "polling" not in (v["reason"] or "") or "no polling" in v["reason"]
+
+
+def test_15m_blocked_falls_back_to_preauthorized():
+    prog = _four_stage_programme()
+    prog.fallbacks = ["fallback-docs"]
+    prog.stages[0].state = COMPLETE
+    prog.stages[1].state = BLOCKED   # provider outage, say
+    prog.stages[2].state = BLOCKED
+    v = prog.continuation_canary()
+    assert v["verdict"] == "PROGRAMME_OPEN"
+    assert v["next_action"] == ("fallback", "fallback-docs"), \
+        "blocked branch must not idle the programme"
+
+
+def test_15m_approval_required_is_honest_stop():
+    prog = _four_stage_programme()
+    prog.stages[0].state = COMPLETE
+    prog.stages[1].state = APPROVAL_REQUIRED
+    prog.stages[2].state = APPROVAL_REQUIRED
+    prog.stages[3].state = APPROVAL_REQUIRED
+    v = prog.continuation_canary()
+    assert v["verdict"] == "APPROVAL_REQUIRED"
+    assert v["next_action"] is None
+
+
+def test_15m_every_stop_carries_an_honest_verdict():
+    """No voluntary stop may be classified UNKNOWN or missing."""
+    states = (PENDING, ELIGIBLE, WAITING_EVENT, COMPLETE, BLOCKED,
+              APPROVAL_REQUIRED, FAILED)
+    for combo_state in states:
+        prog = _four_stage_programme()
+        prog.stages[0].state = COMPLETE
+        for s in prog.stages[1:]:
+            s.state = combo_state
+        v = prog.continuation_canary()
+        assert v["verdict"] in (
+            "PROGRAMME_COMPLETE", "PROGRAMME_OPEN", "PROGRAMME_OPEN_WAITING",
+            "ALL_REMAINING_BLOCKED", "APPROVAL_REQUIRED",
+            "PROGRAMME_COMPLETE_WITH_FAILURES",  # T17 17E: honest all-terminal
+            "UNKNOWN_INTERRUPTION"), v
+        if v["verdict"] == "UNKNOWN_INTERRUPTION":
+            assert combo_state == FAILED, \
+                "UNKNOWN only tolerable for genuinely unclassifiable state"
+
+
+def test_15m_run_programme_drives_without_manual_continue(tmp_path):
+    """End-to-end: the runner drives all 4 stages autonomously; the
+    transition log is the proof artifact; it persists and reloads."""
+    prog = _four_stage_programme()
+
+    def stage_runner(stage):
+        # every stage completes; stage 3's "builder wait" is simulated by
+        # the runner returning COMPLETE (the event mechanism itself is
+        # proven in test_event_wait.py)
+        return COMPLETE
+
+    log = run_programme(prog, stage_runner)
+    verdicts = [e["canary"]["verdict"] for e in log if "canary" in e]
+    assert verdicts.count("PROGRAMME_OPEN") >= 3, \
+        "each completed stage must be followed by an automatic next start"
+    assert verdicts[-1] == "PROGRAMME_COMPLETE"
+    assert prog.stages[3].state == COMPLETE
+
+    # durable: persist + fresh reload → same state, no transcript
+    prog.save(tmp_path)
+    reloaded = Programme.load(tmp_path, "prog-15m-proof")
+    assert reloaded is not None
+    assert all(s.state == COMPLETE for s in reloaded.stages)
+    assert reloaded.continuation_canary()["verdict"] == "PROGRAMME_COMPLETE"
+
+
+def test_15m_dependencies_gate_eligibility():
+    prog = _four_stage_programme()
+    # s2 depends on s1 — with s1 incomplete, s2 is NOT eligible
+    eligible = prog.eligible_stages()
+    assert [s.stage_id for s in eligible] == ["s1"]
+    prog.stages[0].state = COMPLETE
+    eligible = prog.eligible_stages()
+    assert [s.stage_id for s in eligible] == ["s2", "s3", "s4"] \
+        or [s.stage_id for s in eligible] == ["s2"], \
+        "s3/s4 depend transitively on s2/s3 — not eligible yet"

tests/test_programme_wiring.py

@@ -0,0 +1,160 @@
+"""test_programme_wiring.py — T17 §17D/§17E: canonical programme
+continuation wiring held-outs.
+
+The 15M canary proved the RULE; these prove the WIRING: stages bound to
+real Scaffolding contracts, driven through the Construction Interface
+lifecycle, with restart/duplicate/idempotency/adversarial cases.
+"""
+
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune import construction_interface as ci  # noqa: E402
+from rune import programme_continuation as pc  # noqa: E402
+
+
+@pytest.fixture()
+def iface(tmp_path):
+    return ci.ConstructionInterface(
+        builder="t17-programme-test",
+        project_store=str(tmp_path / "scaffold"),
+        project_id="prog-test")
+
+
+def _make_contract(iface, contract_id, need="stage need"):
+    doc = iface.scaffold.load()
+    iface.scaffold.new_contract(doc, contract_id=contract_id, need=need,
+                                origin="programme", affected_entities=[])
+    iface.scaffold.save(doc)
+
+
+def _prog(prog_id, stages):
+    p = pc.Programme(programme_id=prog_id, objective="test programme")
+    for sid, work in stages:
+        st = pc.ProgrammeStage(stage_id=sid, objective=sid,
+                               priority=len(p.stages))
+        if work is not None:
+            st.result["work"] = work
+        p.stages.append(st)
+    return p
+
+
+def _phase(iface, contract_id):
+    doc = iface.scaffold.load()
+    c = iface.scaffold.contract(doc, contract_id)
+    return (c or {}).get("phase", "")
+
+
+def test_17e_all_four_stages_close_with_evidence(iface, tmp_path):
+    stages = [("st-1", None),
+              ("st-2", {"ok": True, "evidence": "did the thing"}),
+              ("st-3", {"ok": True, "evidence": "more work"})]
+    prog = _prog("p-four", stages)
+    for sid, _ in stages:
+        _make_contract(iface, sid)
+    log = pc.run_construction_programme(prog, iface,
+                                        scaffold_store=tmp_path / "scaffold")
+    assert prog.continuation_canary()["verdict"] == "PROGRAMME_COMPLETE"
+    for sid, _ in stages:
+        assert _phase(iface, sid) == "closed"
+    doc = iface.scaffold.load()
+    c = next(e for e in doc["entities"] if e["id"] == "contract:st-2")
+    assert any("did the thing" in (cp.get("note", "") + cp.get("evidence", ""))
+               for cp in c.get("checkpoints", []))
+
+
+def test_17e_failure_does_not_stop_programme(iface, tmp_path):
+    stages = [("f-1", {"ok": False, "evidence": "verification failed"}),
+              ("f-2", {"ok": True, "evidence": "independent work ok"})]
+    prog = _prog("p-fail", stages)
+    for sid, _ in stages:
+        _make_contract(iface, sid)
+    pc.run_construction_programme(prog, iface,
+                                  scaffold_store=tmp_path / "scaffold")
+    verdict = prog.continuation_canary()["verdict"]
+    assert verdict == "PROGRAMME_COMPLETE_WITH_FAILURES"
+    assert _phase(iface, "f-1") == "failed"   # truthful failure, not hung
+    assert _phase(iface, "f-2") == "closed"
+
+
+def test_17e_restart_resumes_from_durable_record(iface, tmp_path):
+    stages = [("r-1", {"ok": True, "evidence": "done"}), ("r-2", None)]
+    prog = _prog("p-restart", stages)
+    for sid, _ in stages:
+        _make_contract(iface, sid)
+    base = tmp_path / "scaffold" / "programmes"
+    # stage 1 genuinely completed before the "restart": close its contract
+    iface.transition("r-1", "approved", evidence_ref="pre")
+    iface.transition("r-1", "building", evidence_ref="pre")
+    iface.transition("r-1", "verifying", evidence_ref="pre")
+    iface.close("r-1", observed_evidence=[{"source": "pre", "detail": "x"}],
+                scope_files=[])
+    prog.stages[0].state = pc.COMPLETE
+    prog.save(base)
+    # "restart": reload from the durable record only
+    prog2 = pc.Programme.load(base, "p-restart")
+    assert prog2.stages[0].state == pc.COMPLETE
+    assert prog2.stages[1].state == pc.PENDING
+    pc.run_construction_programme(prog2, iface,
+                                  scaffold_store=tmp_path / "scaffold")
+    assert _phase(iface, "r-1") == "closed"
+    assert _phase(iface, "r-2") == "closed"
+
+
+def test_17e_duplicate_continuation_signal_idempotent(iface, tmp_path):
+    _make_contract(iface, "d-1")
+    prog = _prog("p-dup", [("d-1", {"ok": True, "evidence": "once"})])
+    base = tmp_path / "scaffold"
+    log1 = pc.run_construction_programme(prog, iface, scaffold_store=base)
+    # re-run the whole programme against the SAME contracts
+    prog2 = pc.Programme.load(base / "programmes", "p-dup")
+    log2 = pc.run_construction_programme(prog2, iface, scaffold_store=base)
+    assert _phase(iface, "d-1") == "closed"   # not re-opened, not corrupted
+    assert any(e["canary"]["verdict"] == "PROGRAMME_COMPLETE"
+               for e in log2 if "canary" in e)
+
+
+def test_17e_already_completed_stage_skipped(iface, tmp_path):
+    _make_contract(iface, "s-1")
+    iface.transition("s-1", "approved", evidence_ref="pre")
+    iface.transition("s-1", "building", evidence_ref="pre")
+    iface.transition("s-1", "verifying", evidence_ref="pre")
+    iface.close("s-1", observed_evidence=[{"source": "pre", "detail": "x"}],
+                scope_files=[])
+    prog = _prog("p-skip", [("s-1", {"ok": True, "evidence": "rework?"})])
+    log = pc.run_construction_programme(prog, iface,
+                                        scaffold_store=tmp_path / "scaffold")
+    # already-closed → COMPLETE without re-running work
+    assert prog.continuation_canary()["verdict"] == "PROGRAMME_COMPLETE"
+    assert not any("rework" in str(e) for e in log)
+
+
+def test_17e_work_callable_receives_interface_and_stage(iface, tmp_path):
+    seen = {}
+
+    def work(i, stage):
+        seen["iface"] = type(i).__name__
+        seen["stage"] = stage.stage_id
+        return {"ok": True, "evidence": "callable ran"}
+
+    _make_contract(iface, "w-1")
+    prog = _prog("p-call", [("w-1", work)])
+    pc.run_construction_programme(prog, iface,
+                                  scaffold_store=tmp_path / "scaffold")
+    assert seen == {"iface": "ConstructionInterface", "stage": "w-1"}
+    assert _phase(iface, "w-1") == "closed"
+
+
+def test_17e_missing_contract_fails_honestly(iface, tmp_path):
+    prog = _prog("p-missing", [("ghost-1", {"ok": True, "evidence": "x"})])
+    log = pc.run_construction_programme(prog, iface,
+                                        scaffold_store=tmp_path / "scaffold")
+    assert prog.stages[0].state == pc.FAILED
+    assert "no contract" in (prog.stages[0].result.get("error") or "")

tests/test_provider_ownership.py

@@ -0,0 +1,198 @@
+"""test_provider_ownership.py — T16 §16F/§16G: provider ownership
+boundary + provider failure held-outs.
+
+THE RULE (permanent architecture):
+    SCAFFOLDING DOES NOT OWN PROVIDER SELECTION.
+
+Proofs:
+  16F guard     — no provider/model name leaks into Scaffolding modules
+                  (structural, permanent — the audit is enforced by the
+                  suite, not by a one-off grep).
+  16F seam      — provider_context.require_model resolves from the ambient
+                  context and fails LOUDLY (ProviderUnresolved) when the
+                  provider layer has not selected a model; no silent
+                  default exists.
+  16G-A/B       — provider context flows THROUGH the construction path:
+                  the builder driver inherits whatever the provider layer
+                  set; Scaffolding never names one.
+  16G-C         — a real provider error (401/403/404/429/5xx) is preserved
+                  VERBATIM — never relabelled a generic "400".
+  16G-D         — NO fallback: a provider failure attempts zero alternate
+                  providers (policy = block_branch_preserve_error, 0
+                  retries).
+  16G-E         — the construction state remains resumable after a
+                  provider block (block ≠ dead contract).
+  16G-G         — Scaffolding canonical project truth is unaffected by
+                  provider preference: switching the provider context does
+                  not mutate scaffold state.
+"""
+
+from __future__ import annotations
+
+import importlib.util
+import sys
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune import failure_recovery as fr  # noqa: E402
+from rune.provider_context import ProviderUnresolved, require_model  # noqa: E402
+
+# The Scaffolding core: construction authority modules that must never
+# own provider selection (T16 §16F audit scope).
+_SCAFFOLDING_CORE = [
+    "scaffold.py", "construction_interface.py",
+    "construction_decision.py", "construction_mode.py",
+    "construction_pattern.py", "construction_patterns_builtin.py",
+    "stage_guards.py", "event_wait.py",
+    "failure_recovery.py", "structural_store.py",
+]
+# construction_* modules discovered live — new ones are covered
+# automatically (the guard follows the surface, not a hand-list)
+import re as _re
+SCAFFOLDING_MODULES = sorted(set(
+    _re.match(r"(construction_\w+)\.py", p.name).group(1) + ".py"
+    for p in (REPO / "src" / "rune").glob("construction_*.py")
+) | set(_SCAFFOLDING_CORE))
+_PROVIDER_TOKENS = [
+    "Verboo", "verboo", "glm-5.3-flash", "mimo-v2.5-pro", "Xioami",
+    "xiaomi", "qwen3:14b", "BUILDER_MODEL", "MIMO_MODEL",
+]
+
+
+def test_16f_no_provider_names_in_scaffolding_modules():
+    """Structural guard: the construction authority is provider-blind."""
+    src = REPO / "src" / "rune"
+    for mod in SCAFFOLDING_MODULES:
+        text = (src / mod).read_text()
+        # strip comments and docstrings crudely but decisively: any REAL
+        # code reference would survive; pure prose mentions live in
+        # comments/docstrings. We assert on code lines only.
+        code_lines = [
+            ln for ln in text.splitlines()
+            if ln.strip() and not ln.strip().startswith(("#", '"', "'"))
+        ]
+        for token in _PROVIDER_TOKENS:
+            hits = [ln for ln in code_lines if token in ln]
+            assert not hits, (
+                f"Scaffolding module {mod} references provider token "
+                f"{token!r} in CODE: {hits[:2]} — Scaffolding must not "
+                f"own provider selection (T16 §16F)")
+
+
+def test_16f_require_model_resolves_ambient_context(monkeypatch):
+    monkeypatch.setenv("MIMO_MODEL", "Provider A/model-a")
+    assert require_model() == "Provider A/model-a"
+
+
+def test_16f_missing_context_fails_loudly_never_silent(monkeypatch):
+    monkeypatch.delenv("MIMO_MODEL", raising=False)
+    with pytest.raises(ProviderUnresolved) as ei:
+        require_model()
+    assert "PROVIDER_UNRESOLVED" in str(ei.value)
+
+
+def test_16g_c_real_provider_status_preserved_verbatim():
+    """No test may pretend all provider failures are a generic '400'."""
+    for err in ("HTTP 401 Unauthorized: bad api key",
+                "HTTP 403 Forbidden: quota tier",
+                "HTTP 404 model not found: glm-5.3-flash",
+                "HTTP 429 rate limited: retry after 30s",
+                "HTTP 502 upstream unavailable"):
+        verdict = fr.classify_failure(err, context={"provider_error": True})
+        assert verdict["class"] == fr.PROVIDER_UNAVAILABLE
+        # the REAL code survives into the evidence, verbatim
+        assert err.split(":")[0] in verdict["evidence"] or \
+            err in verdict["evidence"]
+        assert verdict["policy"] == fr.POLICY[fr.PROVIDER_UNAVAILABLE]
+        assert verdict["policy"][0] == 0  # zero retries, zero fallbacks
+
+
+def test_16g_d_no_fallback_provider_attempted():
+    """A provider failure stops the branch — the plan of record contains
+    exactly ONE provider attempt, and the recovery policy blocks."""
+    attempted_providers: list[str] = []
+
+    def construction_path(provider_ctx: str, fail_on: str):
+        attempted_providers.append(provider_ctx)
+        try:
+            if provider_ctx == fail_on:
+                raise RuntimeError(f"HTTP 401 Unauthorized from {provider_ctx}")
+        except RuntimeError as exc:
+            verdict = fr.classify_failure(str(exc),
+                                          context={"provider_error": True})
+            if verdict["policy"][0] == 0:  # block_branch_preserve_error
+                return verdict  # STOP the dependent branch — no plan B
+        return None
+
+    verdict = construction_path("Provider A/model-a", "Provider A/model-a")
+    assert attempted_providers == ["Provider A/model-a"]  # 16G-D: ONE
+    assert verdict["class"] == fr.PROVIDER_UNAVAILABLE
+    assert "401" in verdict["evidence"]  # real code preserved
+
+
+def test_16g_e_construction_remains_resumable_after_provider_block(tmp_path):
+    """Block ≠ dead: after a provider-unavailable block the scaffold
+    contract must still be loadable and non-terminal (resume stays
+    legal)."""
+    from rune import failure_recovery as fr
+    from rune import scaffold as sc
+    s = sc.Scaffold(tmp_path / ".rune" / "scaffold", "prov-e")
+    doc = s.init(purpose="provider block resumability")
+    s.new_contract(doc, contract_id="prov-e-1", need="build a thing")
+    verdict = fr.classify_failure("HTTP 429 quota exhausted",
+                                  context={"provider_error": True})
+    for e in doc["entities"]:
+        if e["id"] == "contract:prov-e-1":
+            e["identities"]["blocker"] = {
+                "class": verdict["class"], "policy": verdict["policy"],
+                "evidence": verdict["evidence"],
+            }
+    s.save(doc)
+    # a FRESH process reads the same contract: non-terminal + resumable
+    doc2 = sc.Scaffold(tmp_path / ".rune" / "scaffold", "prov-e").load()
+    c = next(e for e in doc2["entities"]
+             if e["id"] == "contract:prov-e-1")
+    assert c["identities"]["blocker"]["class"] == "PROVIDER_UNAVAILABLE"
+    assert c["identities"]["blocker"]["policy"][1].startswith("block_branch")
+    assert c.get("phase") != "closed"  # block is NOT completion
+
+
+def test_16g_g_provider_preference_never_touches_scaffold_truth(tmp_path):
+    """Switching the provider context must leave Scaffolding canonical
+    project state byte-identical."""
+    import os
+    from rune import scaffold as sc
+    s = sc.Scaffold(tmp_path / ".rune" / "scaffold", "prov-g")
+    doc = s.init(purpose="provider neutrality")
+    s.new_contract(doc, contract_id="prov-g-1", need="a need")
+    s.save(doc)
+    before = (tmp_path / ".rune" / "scaffold" / "current.json").read_text()
+    for ctx in ("Provider A/model-a", "Provider B/model-b", None):
+        if ctx:
+            os.environ["MIMO_MODEL"] = ctx
+        else:
+            os.environ.pop("MIMO_MODEL", None)
+        s2 = sc.Scaffold(tmp_path / ".rune" / "scaffold", "prov-g")
+        doc2 = s2.init(purpose="provider neutrality")  # idempotent re-init
+        s2.save(doc2)
+        after = (tmp_path / ".rune" / "scaffold" / "current.json").read_text()
+        assert "Provider" not in after  # provider context never leaks in
+        before = after
+    os.environ.pop("MIMO_MODEL", None)
+
+
+def test_16g_drivers_have_no_silent_provider_default():
+    """The builder/research drivers resolve provider via require_model —
+    a missing context exits 3 (PROVIDER_UNRESOLVED), never a baked-in
+    provider name."""
+    for script in ("t14_builder_session.py", "t15_builder_session.py"):
+        text = (REPO / "scripts" / script).read_text()
+        assert "require_model" in text, f"{script} bypasses the seam"
+        assert "Verboo Code/glm-5.3-flash\"\" )" not in text
+        assert '"Verboo Code/glm-5.3-flash")' not in text and \
+            "'Verboo Code/glm-5.3-flash')" not in text, (
+            f"{script} still carries a hardcoded provider default")

tests/test_stage_guards.py

@@ -0,0 +1,216 @@
+"""test_stage_guards.py — T15 §15F/§15G/§15H/§15I held-outs.
+
+Hostile-first: these tests attempt to falsify the guard semantics,
+not encode them.
+
+15F  readiness failure BLOCKS the action (action must NOT run)
+15F  exit canary failure prevents stage PASS even when the action
+     "succeeded" (self-certification impossible)
+15F  liveness: inactivity is STALLED, never completion
+15G  zero tests + exit 0 → FALSE_GREEN_REFUSED
+15G  all-skipped + exit 0 → refused
+15G  contract-required artifact absent → refused despite green command
+15G  genuine pass with contract-sufficient tests → PASS
+15H  Henxels REQUIRED at structural boundaries / FULL depth;
+     NOT_APPLICABLE for tiny bounded changes — always with rationale
+15I  command exits 0 but durable effect absent → delta non-zero →
+     the next stage MUST NOT start
+15I  reconcile zero-delta → continue
+"""
+
+import json
+import subprocess
+import subprocess
+import subprocess
+import os
+import sys
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune.stage_guards import (  # noqa: E402
+    ExitCanaryFailed, StageBlocked, StageContract, StageCheck,
+    TestExpectations, execute_stage, henxels_placement, liveness_canary,
+    readiness_canary, reconcile, verify_test_truth,
+)
+
+
+def _stage(tmp_path, *, ready=None, done=None):
+    marker = tmp_path / "effect.txt"
+    return StageContract(
+        stage_id="s1", contract_id="contract:t15",
+        ready_when=ready if ready is not None else [],
+        done_when=done if done is not None else [
+            StageCheck(name="effect", kind="path_exists", path=str(marker))],
+        expected_effect="marker exists",
+    ), marker
+
+
+# ── 15F ─────────────────────────────────────────────────────────────────────
+
+def test_15f_readiness_failure_blocks_action(tmp_path):
+    sentry = tmp_path / "action-ran.txt"
+    stage, _ = _stage(tmp_path, ready=[
+        StageCheck(name="prereq", kind="path_exists",
+                   path=str(tmp_path / "missing-prereq.txt"))])
+    def action():
+        sentry.write_text("must not happen")
+        return {}
+    with pytest.raises(StageBlocked) as e:
+        execute_stage(stage, action)
+    assert e.value.failures[0]["name"] == "prereq"
+    assert not sentry.exists(), "blocked stage must NOT execute"
+
+
+def test_15f_exit_canary_refuses_self_certification(tmp_path):
+    """The action 'succeeds' and CLAIMS completion — but the expected
+    effect does not exist in reality. Exit canary refuses."""
+    stage, marker = _stage(tmp_path)  # done_when: marker must exist
+    def lying_action():
+        return {"claimed": "complete, trust me"}
+    with pytest.raises(ExitCanaryFailed) as e:
+        execute_stage(stage, lying_action)
+    assert e.value.failures[0]["name"] == "effect"
+    assert not marker.exists()
+
+
+def test_15f_exit_canary_passes_on_real_effect(tmp_path):
+    stage, marker = _stage(tmp_path)
+    def real_action():
+        marker.write_text("done")
+        return {"wrote": str(marker)}
+    result = execute_stage(stage, real_action)
+    assert result["verdict"] == "PASS"
+
+
+def test_15f_liveness_inactivity_is_not_completion():
+    alive = liveness_canary(last_progress_at=100.0, now=110.0)
+    stalled = liveness_canary(last_progress_at=100.0, now=500.0)
+    assert alive["verdict"] == "ALIVE"
+    assert stalled["verdict"] == "STALLED"
+    assert stalled["stalled"] is True
+    # a stalled stage has NO completion semantics — only STALLED
+    assert "complete" not in stalled["verdict"].lower()
+
+
+# ── 15G ─────────────────────────────────────────────────────────────────────
+
+def _write_node_project(tmp_path, body: str):
+    (tmp_path / "scheduler.test.js").write_text(body)
+
+
+def test_15g_zero_tests_with_green_exit_is_false_green(tmp_path):
+    _write_node_project(tmp_path, "console.log('nothing tested');")
+    r = verify_test_truth(
+        ["node", "scheduler.test.js"], str(tmp_path),
+        TestExpectations(require_tests=True), base_dir=str(tmp_path))
+    assert r["verdict"] == "FALSE_GREEN_REFUSED"
+    assert any(f["name"] == "test_count" for f in r["failures"])
+
+
+def test_15g_all_skipped_is_refused(tmp_path):
+    # 2 "tests" all skipped: node --test reports pass 0 skip 2 → the
+    # zero-count guard also triggers; craft a runner-shaped output where
+    # pass counts exist but all runs were skips is covered by the
+    # skip-ratio check via a fake runner script
+    runner = tmp_path / "fake-runner.sh"
+    runner.write_text('#!/bin/bash\necho "# pass 5"\necho "# skipped 5"\nexit 0\n')
+    runner.chmod(0o755)
+    r = verify_test_truth([str(runner)], str(tmp_path),
+                          TestExpectations(require_tests=True),
+                          base_dir=str(tmp_path))
+    assert r["verdict"] == "FALSE_GREEN_REFUSED"
+    assert any(f["name"] == "all_skipped" for f in r["failures"])
+
+
+def test_15g_missing_required_artifact_refused_despite_green(tmp_path):
+    runner = tmp_path / "green.sh"
+    runner.write_text('#!/bin/bash\necho "# pass 11"\nexit 0\n')
+    runner.chmod(0o755)
+    r = verify_test_truth([str(runner)], str(tmp_path),
+                          TestExpectations(
+                              require_tests=True,
+                              required_artifacts=["scheduler.js"]),
+                          base_dir=str(tmp_path))
+    assert r["verdict"] == "FALSE_GREEN_REFUSED"
+    assert any(f["name"] == "artifact_missing" for f in r["failures"])
+
+
+def test_15g_genuine_pass_with_real_tests(tmp_path):
+    _write_node_project(tmp_path, (
+        "import test from 'node:test';\n"
+        "import assert from 'node:assert/strict';\n"
+        "test('works', () => assert.equal(1+1, 2));\n"
+        "test('also', () => assert.ok(true));\n"))
+    (tmp_path / "scheduler.js").write_text("export const x = 1\n")
+    r = verify_test_truth(
+        ["node", "--test"], str(tmp_path),
+        TestExpectations(require_tests=True, min_tests=2,
+                         required_artifacts=["scheduler.js"]),
+        base_dir=str(tmp_path))
+    assert r["verdict"] == "PASS", r["failures"]
+    assert r["test_count"] >= 2
+
+
+def test_15g_expectations_come_from_contract_not_hardcoded(tmp_path):
+    """min_tests is contract-derived: a contract may legitimately require
+    more; the same green output fails a higher requirement."""
+    runner = tmp_path / "g.sh"
+    runner.write_text('#!/bin/bash\necho "# pass 3"\nexit 0\n')
+    runner.chmod(0o755)
+    ok = verify_test_truth([str(runner)], str(tmp_path),
+                           TestExpectations(min_tests=3), base_dir=str(tmp_path))
+    strict = verify_test_truth([str(runner)], str(tmp_path),
+                               TestExpectations(min_tests=11), base_dir=str(tmp_path))
+    assert ok["verdict"] == "PASS"
+    assert strict["verdict"] == "FALSE_GREEN_REFUSED"
+
+
+# ── 15H ─────────────────────────────────────────────────────────────────────
+
+def test_15h_henxels_required_at_structural_boundary():
+    r = henxels_placement(depth="STANDARD", structural_boundary=True,
+                          stage_id="s-structural")
+    assert r["henxels"] == "HENXELS_REQUIRED"
+    assert r["rationale"]
+
+
+def test_15h_henxels_required_for_full_depth():
+    r = henxels_placement(depth="FULL", structural_boundary=False,
+                          stage_id="s-full")
+    assert r["henxels"] == "HENXELS_REQUIRED"
+
+
+def test_15h_henxels_not_applicable_for_tiny_change():
+    r = henxels_placement(depth="LIGHT", structural_boundary=False,
+                          stage_id="s-typo")
+    assert r["henxels"] == "HENXELS_NOT_APPLICABLE"
+    assert r["rationale"]  # NOT_APPLICABLE is a recorded decision too
+
+
+# ── 15I ─────────────────────────────────────────────────────────────────────
+
+def test_15i_command_green_but_effect_absent_blocks_next_stage(tmp_path):
+    """The classic lie: subprocess exits 0, durable effect absent.
+    Reconciliation computes the delta; the next stage must not start."""
+    desired = {"deployed_marker": str(tmp_path / "deployed.txt")}
+    # "action" that exits 0 but does nothing durable:
+    r = subprocess.run(["bash", "-c", "exit 0"], capture_output=True)
+    assert r.returncode == 0  # the command "succeeded"
+    delta = reconcile({"deployed_marker": "present"},
+                      lambda: {"deployed_marker": "absent"},
+                      stage_id="deploy")
+    assert delta["reconciled"] is False
+    assert delta["delta"]["deployed_marker"]["desired"] == "present"
+    assert delta["delta"]["deployed_marker"]["actual"] == "absent"
+
+
+def test_15i_reconciled_zero_delta_continues(tmp_path):
+    actual = {"phase": "closed", "scope": ["a.js"]}
+    r = reconcile({"phase": "closed", "scope": ["a.js"]}, lambda: actual,
+                  stage_id="close")
+    assert r["reconciled"] is True
+    assert r["delta"] == {}

tests/test_substrate_seam.py

@@ -0,0 +1,176 @@
+"""test_substrate_seam.py — T16 §16I/§16J/§16K: future-Substrate
+compatibility seam held-outs.
+
+The MOCK client below knows NOTHING about current.json or internal
+entity arrays — it sees only the public projection (construction_
+projection) and the inbound command seam (construction_commands). It:
+
+  1. requests current construction state through the public seam;
+  2. observes a construction event (a projection whose waiting_state
+     advanced — simulated by a durable state change, not by poking
+     internals);
+  3. correlates project/contract/stage;
+  4. issues RESUME (valid) and receives acceptance;
+  5. issues an invalid transition (START on a non-draft) and receives
+     REFUSAL;
+  6. survives process restart using identities (re-request state by
+     contract id alone — no transcript, no cached internals);
+  7. leaves Scaffolding fully usable without the client existing.
+
+NOT implemented (explicitly out of scope): relevance, autonomous
+prioritisation, goal planning, memory, Substrate cognition. This is a
+compatibility seam, not a Substrate.
+"""
+
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+import pytest
+
+REPO = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(REPO / "src"))
+
+from rune import construction_commands as cmds  # noqa: E402
+from rune import construction_projection as proj  # noqa: E402
+from rune import scaffold as sc  # noqa: E402
+
+
+class MockSubstrateClient:
+    """TEST CLIENT ONLY. Sees the projection + command seam; the store
+    object is passed to the SEAM functions, never touched directly."""
+
+    def __init__(self, scaffold):
+        self._scaffold = scaffold
+        self.observed: list[dict] = []
+
+    def request_state(self, contract_id: str, **kw) -> dict:
+        p = proj.project(self._scaffold, contract_id, **kw)
+        assert not proj.validate(p), f"invalid projection: {proj.validate(p)}"
+        self.observed.append(p)
+        return p
+
+    def issue(self, command_type: str, construction_id: str, **kw) -> dict:
+        # the serving layer binds the canonical store to the seam; the
+        # client only ever sees the returned envelope
+        return cmds.command(command_type, construction_id=construction_id,
+                            scaffold=self._scaffold, **kw)
+
+
+def _fixture_scaffold(tmp_path):
+    s = sc.Scaffold(tmp_path / ".rune" / "scaffold", "substrate-seam")
+    doc = s.init(purpose="substrate seam held-out")
+    s.new_contract(doc, contract_id="seam-1", need="a thing to build")
+    s.save(doc)
+    return s
+
+
+def test_16k_state_through_public_seam_only(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    p = client.request_state("seam-1", originating_goal_id="goal-9",
+                             stage="BUILD")
+    assert p["schema_version"] == proj.SCHEMA_VERSION
+    assert p["contract_id"] == "seam-1"
+    assert p["phase"] == "draft"
+    assert p["candidate_next_action"]  # Scaffolding tells, client reads
+    # the projection exposes NO storage internals
+    assert "entities" not in p and "current.json" not in str(p)
+
+
+def test_16k_observe_event_correlate_and_resume(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    before = client.request_state("seam-1", stage="BUILD")
+    # a construction EVENT: durable state advanced (transition to
+    # approved — done through the real lifecycle, not internals)
+    doc = s.load()
+    s.transition_contract(doc, "seam-1", "approved")
+    s.save(doc)
+    after = client.request_state("seam-1", stage="BUILD")
+    assert before["phase"] == "draft" and after["phase"] == "approved"
+    # correlate project/contract/stage, then issue the valid command
+    assert after["project_id"] == "substrate-seam"
+    resp = client.issue("RESUME", construction_id="seam-1",
+                        correlation_id="corr-1")
+    assert resp["accepted"] is True and resp["op"] == "resume"
+
+
+def test_16k_invalid_transition_refused(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    doc = s.load()
+    s.transition_contract(doc, "seam-1", "approved")
+    s.save(doc)
+    # START on a non-draft is invalid — Scaffolding decides, Substrate asks
+    resp = client.issue("START", construction_id="seam-1")
+    assert resp["accepted"] is False
+    assert "draft" in resp["reason"]
+
+
+def test_16k_direct_store_semantics_impossible(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    # the client has no path to the store: it never received the store
+    # dir, only the scaffold object for seam calls. Attempting to mutate
+    # storage is not expressible through the seam — the closest it gets
+    # is PROVIDE_CONTEXT, which is validated and user-provenance-gated.
+    resp = client.issue("PROVIDE_CONTEXT", construction_id="seam-1",
+                        payload={"add_paths": ["src/x.py"]})
+    assert resp["accepted"] is True
+    assert resp["args"]["authority"] == "substrate"  # never user-forged
+
+
+def test_16k_survives_restart_via_identities(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    # "restart": drop the client AND the scaffold; rebuild from identities
+    contract_id, project_id = "seam-1", "substrate-seam"
+    del s
+    s2 = sc.Scaffold(tmp_path / ".rune" / "scaffold", project_id)
+    client2 = MockSubstrateClient(s2)
+    p = client2.request_state(contract_id)  # contract id alone — no transcript
+    assert p["phase"] == "draft" and p["project_id"] == project_id
+
+
+def test_16k_scaffolding_usable_without_client(tmp_path):
+    # the client never existed here: the full lifecycle still works
+    s = _fixture_scaffold(tmp_path)
+    doc = s.load()
+    s.transition_contract(doc, "seam-1", "approved")
+    s.transition_contract(doc, "seam-1", "building")
+    s.save(doc)
+    doc2 = s.load()
+    c = next(e for e in doc2["entities"] if e["id"] == "contract:seam-1")
+    assert c["phase"] == "building"
+
+
+def test_16k_idempotent_command_replay(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    first = client.issue("RESUME", construction_id="seam-1",
+                         command_id="replay-1")
+    second = client.issue("RESUME", construction_id="seam-1",
+                          command_id="replay-1")
+    assert first["accepted"] and second["accepted"]
+    assert second["idempotent_replay"] is True
+
+
+def test_16k_unknown_command_and_unknown_construction_refused(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    assert client.issue("PAUSE", construction_id="seam-1")["accepted"] is False
+    assert client.issue("RESUME", construction_id="nope")["accepted"] is False
+    assert client.issue("RESUME", construction_id="")["accepted"] is False
+
+
+def test_16k_projection_schema_validated(tmp_path):
+    s = _fixture_scaffold(tmp_path)
+    client = MockSubstrateClient(s)
+    p = client.request_state("seam-1")
+    assert proj.validate(p) == []
+    bad = dict(p)
+    del bad["phase"]
+    assert proj.validate(bad) != []
+    bad2 = dict(p, schema_version=999)
+    assert proj.validate(bad2) != []

tests/test_t14_builder_contract.py

@@ -249,3 +249,270 @@ def test_14f_construction_kernel_shared_rune_internal_vs_external(tmp_path):
     # ALLOWED difference: authority identity
     assert ext.scaffold.project_id == "t14-hostile"
     assert rune_iface.scaffold.project_id == "rune"
+
+
+# ── 15O: residue provenance hardening ───────────────────────────────────────
+
+def test_15o_residue_lookalike_malicious_write_is_caught(tmp_path):
+    """§15O hostile: a builder deliberately writes an exempt-looking path
+    (.mimocode/evil, .git/config). Provenance — NOT the prefix — decides:
+    the path appeared mid-session and was never declared by the runtime,
+    so it is a SCOPE VIOLATION, not residue."""
+    sys.path.insert(0, str(ROOT / "scripts"))
+    import t14_builder_session as t14
+    proj = tmp_path / "proj"
+    proj.mkdir()
+    # runtime bootstrap BEFORE builder: .mimocode session dir exists
+    (proj / ".mimocode").mkdir()
+    (proj / ".mimocode" / "session.json").write_text("{}")
+    before = t14.snapshot_files(proj)
+    # builder writes its legit file AND a residue-lookalike malicious one
+    (proj / "scheduler.js").write_text("ok")
+    (proj / ".mimocode" / "evil").write_text("payload")
+    (proj / ".git" / "config").parent.mkdir()
+    (proj / ".git" / "config").write_text("tampered")
+    audit = t14.audit_scope(proj, before, ["scheduler.js"])
+    assert set(audit["scope_violations"]) == {".mimocode/evil", ".git/config"}
+    assert not audit["in_scope"]
+    assert audit["runtime_residue"] == [".mimocode/session.json"] or \
+        ".mimocode/session.json" not in audit["scope_violations"]
+
+
+def test_15o_preexisting_residue_stays_exempt(tmp_path):
+    """Residue with proven provenance (existed before the builder ran, or
+    explicitly runtime-declared) remains exempt and is recorded."""
+    sys.path.insert(0, str(ROOT / "scripts"))
+    import t14_builder_session as t14
+    proj = tmp_path / "proj"
+    (proj / ".mimocode").mkdir(parents=True)
+    (proj / ".mimocode" / "session.json").write_text("{}")
+    before = t14.snapshot_files(proj)
+    (proj / "scheduler.js").write_text("ok")
+    (proj / ".mimocode" / "new-runtime-file").write_text("x")
+    audit = t14.audit_scope(
+        proj, before, ["scheduler.js"],
+        runtime_declared={".mimocode/new-runtime-file"})
+    assert audit["in_scope"]
+    assert ".mimocode/new-runtime-file" in audit["runtime_residue"]
+    assert audit["scope_violations"] == []
+
+
+# ── 15P: null-action / idempotency proofs ───────────────────────────────────
+
+def test_15p_project_build_bootstrap_twice_no_duplicate(tmp_path):
+    """15P: PROJECT_BUILD bootstrap executed twice must be a
+    NOOP_ALREADY_SATISFIED-equivalent — one project identity, not two."""
+    from rune import construction_activation as ca
+    from rune import construction_mode as cm
+    store = tmp_path / "scaffold"
+    s = sc.Scaffold(str(store), "rune")
+    s.init(purpose="15p probe")
+    d1 = cm.classify_mode(
+        "create a standalone practice scheduler with create/list/complete",
+        target_project=None, project_exists=False,
+        affected_surface=["scheduler.js"])
+    from rune.construction_decision import Need
+    need = Need(need_id="n1", objective="create a standalone practice "
+                "scheduler with create/list/complete", origin="test",
+                project="scheduler", constraints={}, affected_surface=[],
+                evidence_refs=[], approval_context="15p")
+    r1 = ca.activate(d1, need=need, scaffold=s, builder="probe")
+    r2 = ca.activate(d1, need=need, scaffold=s, builder="probe")
+    assert r1["mode"] == "PROJECT_BUILD" and r2["mode"] == "PROJECT_BUILD"
+    # exactly ONE project scaffold — no duplicate identity
+    projects = sorted((store / "projects").glob("*/current.json"))
+    assert len(projects) == 1, f"bootstrap duplicated: {projects}"
+
+
+def test_15p_event_replay_after_restart_is_idempotent():
+    """15P + 15E: replaying a consumed operation event after a controller
+    restart must be a no-op — the durable waitstore verdict stands."""
+    from rune import event_wait as ew
+    # (full daemon-backed proof lives in test_event_wait.py
+    #  test_15e_restart_after_event_persisted_waitstore_wins; this is the
+    #  semantic contract: resume_or_recover is a pure readback)
+    # the function never mutates state — verify by contract inspection
+    import inspect
+    src = inspect.getsource(ew.resume_or_recover)
+    assert "save(" not in src and "transition(" not in src, \
+        "recovery must be readback-only (idempotent by construction)"
+
+
+# ── 15S: breaking-change / migration scenario ───────────────────────────────
+
+def test_15s_breaking_change_preserves_old_path_until_migrated(tmp_path):
+    """15S: existing project → breaking change → migration requirement →
+    FULL depth. BUILD NEW does not destroy the old path before migration
+    evidence exists; canaries block supersession until consumers migrate."""
+    from rune import construction_activation as ca
+    from rune import construction_mode as cm
+    store = tmp_path / "scaffold"
+    s = sc.Scaffold(str(store), "rune")
+    s.init(purpose="15s probe")
+    doc = s.load()
+    # existing project with a consumer of the old API
+    ss.add_entity(doc, id="module:v1_api", kind="module",
+                  identities={"api": "get_items()"}, maturity="WIRED",
+                  evidence_refs=["e"])
+    ss.add_entity(doc, id="consumer:dashboard", kind="capability",
+                  identities={"consumes": "module:v1_api"},
+                  maturity="WIRED", evidence_refs=["e"])
+    ss.add_relationship(doc, type="consumes", from_id="consumer:dashboard",
+                        to_id="module:v1_api", evidence_refs=["e"])
+    ss.add_entity(doc, id="contract:api-spec", kind="contract",
+                  identities={"objective": "api spec"}, evidence_refs=["e"])
+    ss.add_entity(doc, id="capability:api-v2", kind="capability",
+                  maturity="DECLARED", evidence_refs=["e"])
+    ss.add_relationship(doc, type="implements",
+                        from_id="capability:api-v2",
+                        to_id="contract:api-spec", evidence_refs=["e"])
+    s.commit(doc, what="seed 15s existing project")
+    iface = ci.ConstructionInterface(
+        builder="probe", project_store=str(store), project_id="rune")
+    sub = iface.submit_need(
+        "replace get_items() with get_items_paged() and migrate the "
+        "dashboard consumer",
+        constraints={"environment": {"os": "linux"},
+                     "contracts": ["contract:api-spec"]},
+        affected_surface=["api.py", "dashboard.py"],
+        evidence_refs=["e"], approval_context="15s")
+    assert sub["success"], sub["errors"]
+    cid = sub["ids"]["contract_id"]
+    doc = iface.scaffold.load()
+    c = iface.scaffold.contract(doc, cid)
+    # the old path is IDENTIFIED as an affected entity — governance sees
+    # the consumer before any change happens
+    assert c is not None
+    # transition to verifying, then migrating: supersession gated on
+    # migration evidence — a close from verifying WITHOUT the migrating
+    # phase must still be possible only with verification evidence;
+    # the phase map enforces migrating before superseding for
+    # migration-carrying contracts
+    assert iface.transition(cid, "approved")["success"]
+    assert iface.transition(cid, "building")["success"]
+    assert iface.transition(cid, "verifying")["success"]
+    # BOTH behaviors must be proven: old consumer still works AND new
+    # API exists — the contract carries migration obligations
+    mig = iface.transition(cid, "migrating", evidence_ref="consumer migration")
+    assert mig["success"], mig.get("errors")
+    sup = iface.transition(cid, "superseding", evidence_ref="old path retired")
+    assert sup["success"], sup.get("errors")
+    # the OLD path record survives in the governed store (history, not
+    # deletion): supersession is a lifecycle state, not erasure
+    doc = iface.scaffold.load()
+    assert ss.entity(doc, "module:v1_api") is not None
+    assert ss.entity(doc, "consumer:dashboard") is not None
+
+
+# ── 15T: failure / recovery matrix ──────────────────────────────────────────
+
+def test_15t_failure_classification_matrix():
+    """15T: every failure class classifies deterministically, with an
+    explicit retry budget — and PERMANENT/APPROVAL never auto-retry."""
+    from rune import failure_recovery as fr
+    cases = [
+        ("connection reset by peer", fr.TRANSIENT, 2),
+        ("rate limit 429 exceeded", fr.TRANSIENT, 2),
+        ("test is flaky, occasionally fails", fr.INTERMITTENT, 3),
+        ("schema-invalid observation payload", fr.PERMANENT, 0),
+        ("module not found: foo", fr.PERMANENT, 0),
+        ("user confirmation required to proceed", fr.APPROVAL_REQUIRED, 0),
+        ("destructive external change", fr.APPROVAL_REQUIRED, 0),
+        ("stale document committed over persisted state", fr.KNOWN_PROCESS_DEFECT, 1),
+        ("totally novel failure mode", fr.UNKNOWN, 1),
+    ]
+    for text, expected_class, expected_budget in cases:
+        r = fr.classify_failure(text)
+        assert r["class"] == expected_class, f"{text!r} → {r['class']}"
+        max_retries, action = r["policy"]
+        assert max_retries == expected_budget, text
+    # no infinite loops: every policy has an explicit budget
+    for cls in fr.CLASSES:
+        budget, _ = fr.POLICY[cls]
+        assert isinstance(budget, int) and budget <= 3
+    # PERMANENT and APPROVAL never retry
+    assert fr.POLICY[fr.PERMANENT][0] == 0
+    assert fr.POLICY[fr.APPROVAL_REQUIRED][0] == 0
+
+
+def test_15t_known_defect_context_wins_over_markers():
+    from rune import failure_recovery as fr
+    r = fr.classify_failure("timeout during deploy",
+                            context={"known_defect_id": "process-defect:x"})
+    assert r["class"] == fr.KNOWN_PROCESS_DEFECT
+    assert "process-defect:x" in r["evidence"]
+
+
+# ── 15Y: adversarial default-mechanism matrix (remaining gaps) ──────────────
+
+def test_15y_event_for_different_project_matches_nothing():
+    """15Y case 8: an event for a different project cannot mutate the
+    target wait — correlation is (provider, event_type, run_ref)."""
+    import sys
+    sys.path.insert(0, str(ROOT / "src"))
+    from rune import event_wait as ew
+    import tempfile
+    from pathlib import Path
+    tmp = Path(tempfile.mkdtemp())
+    import os
+    os.environ["RUNE_WR_ROOT"] = str(tmp)
+    os.environ["RUNE_WR_CODE"] = "/home/nick/.local/share/mimo-wait-resume"
+    ew.register_process_wait("op-target", project="project-a",
+                             watchdog_secs=60)
+    ew.register_process_wait("op-other", project="project-b",
+                             watchdog_secs=60)
+    import importlib
+    sys.path.insert(0, "/home/nick/.local/share/mimo-wait-resume")
+    import waitstore
+    waitstore = importlib.reload(waitstore)
+    # the target's correlation is op-target; an event for op-other has a
+    # different run_ref → match() rejects it for the target's wait
+    target = None
+    for rec in waitstore.all_waits():
+        if rec.get("ext", {}).get("operation_id") == "op-target":
+            target = rec
+    assert target is not None
+    foreign_event = {"provider": "builder", "event_type": "builder_exit",
+                     "run_ref": "op-other", "exit_code": 0}
+    assert waitstore.match(target, foreign_event) is False
+    own_event = {"provider": "builder", "event_type": "builder_exit",
+                 "run_ref": "op-target", "exit_code": 0}
+    assert waitstore.match(target, own_event) is True
+
+
+def test_15y_user_approval_gate_blocks_consequential_action():
+    """15Y case 22: a consequential scope change without user authority
+    is refused — amend_scope is user-gated, not builder-gated."""
+    from rune import construction_interface as ci
+    import tempfile
+    from pathlib import Path
+    tmp = Path(tempfile.mkdtemp())
+    iface = ci.ConstructionInterface(builder="rogue-builder",
+                                     project_store=str(tmp / "s"),
+                                     project_id="t15y")
+    iface.scaffold.init(purpose="15y probe")
+    doc = iface.scaffold.load()
+    from rune import structural_store as ss
+    ss.add_entity(doc, id="contract:spec-15y", kind="contract",
+                  identities={"objective": "spec"}, evidence_refs=["e"])
+    ss.add_entity(doc, id="capability:planned", kind="capability",
+                  maturity="DECLARED", evidence_refs=["e"])
+    ss.add_relationship(doc, type="implements",
+                        from_id="capability:planned",
+                        to_id="contract:spec-15y", evidence_refs=["e"])
+    iface.scaffold.commit(doc, what="seed 15y spec")
+    sub = iface.submit_need(
+        "build the 15y module",
+        constraints={"environment": {"os": "linux"},
+                     "contracts": ["contract:spec-15y"]},
+        evidence_refs=["e"], approval_context="15y")
+    assert sub["success"]
+    cid = sub["ids"]["contract_id"]
+    # builder attempts to broaden its own scope — refused
+    r = iface.amend_scope(cid, ["extra/secret.py"], reason="builder wants",
+                          authority="builder")
+    assert not r["success"]
+    # user authority works
+    r2 = iface.amend_scope(cid, ["extra/needed.py"], reason="user asked",
+                           authority="user")
+    assert r2["success"], r2.get("errors")