conductor/rune
1
2 Fork 1
Code Issues 36 Pull Requests 32 Actions 29 Packages Projects Releases 564 Wiki Activity

RP8 (child of epic #257). DEPENDS ON: RP7. BLOCKS: RP10. ACCEPTANCE: - Deterministic-first readiness checks; PROCEED or REFINE with reason, return_to (earliest deficient stage), typed_unknowns, required_fix. - Targeted refinement only — no full-pipe… #271

New Issue
Closed
opened 2026-09-08 23:36:12 +01:00 by conductor · 4 comments
conductor commented 2026-09-08 23:36:12 +01:00
Owner
Copy link
Copy source
No description provided.
conductor commented 2026-09-08 23:36:59 +01:00
Author
Owner
Copy link
Copy source

Parent epic: #257 (Research Runtime P0 — canonical pipeline amendment). Canonical handoff 2026-09-08.

Parent epic: #257 (Research Runtime P0 — canonical pipeline amendment). Canonical handoff 2026-09-08.
gitea-actions Bot commented 2026-09-08 23:37:40 +01:00
Copy link
Copy source

⚠️ Priority gate: this issue has no priority. Assign exactly one of P0 — critical (spine/correctness/blocks), P1 — high (material value or real defect), P2 — medium (quality/consolidation), P3 — low (measurement/deferred). An issue without a priority is not filed.

⚠️ **Priority gate:** this issue has no priority. Assign exactly one of **P0 — critical** (spine/correctness/blocks), **P1 — high** (material value or real defect), **P2 — medium** (quality/consolidation), **P3 — low** (measurement/deferred). An issue without a priority is not filed.
gitea-actions Bot added the bug label 2026-09-08 23:37:46 +01:00
conductor commented 2026-09-09 01:03:36 +01:00
Author
Owner
Copy link
Copy source

Pipeline correction amendment (2026-09-09) — QUALIFY narrowed

QUALIFY is the IMPLEMENTATION-READINESS gate, not a second research loop (§8–10).

Normal outcomes: PROCEED | REFINE_SYNTHESIS (→ SYNTHESIS only; never OUTWARD).
Exceptional: REOPEN_EARLIER_STAGE — requires reason, return_to, new_blocking_fact, why_previous_gate_could_not_reasonably_have_caught_it, required_fix.

Checks: chosen design explicit, interfaces concrete, authority boundaries explicit, data/event flow specified, sequence concrete, invariants defined, tests specified, acceptance observable, failure modes handled, rollback defined, adversarial mitigations included, falsification tests included, blocking implementation unknowns absent.

Generic end-of-pipeline MORE_RESEARCH is REMOVED as an outcome (§11). Frequent reopening of earlier stages is evidence of weak earlier gates and must be treated as a process defect.

Implemented in a553fb1 (QUALIFY_OUTCOMES, _route_after_qualify). Dependency edges: QUALIFY→SYNTHESIS refinement; exceptional reopen→earliest affected stage only.

**Pipeline correction amendment (2026-09-09) — QUALIFY narrowed** QUALIFY is the IMPLEMENTATION-READINESS gate, not a second research loop (§8–10). Normal outcomes: `PROCEED` | `REFINE_SYNTHESIS` (→ SYNTHESIS only; never OUTWARD). Exceptional: `REOPEN_EARLIER_STAGE` — requires reason, return_to, new_blocking_fact, why_previous_gate_could_not_reasonably_have_caught_it, required_fix. Checks: chosen design explicit, interfaces concrete, authority boundaries explicit, data/event flow specified, sequence concrete, invariants defined, tests specified, acceptance observable, failure modes handled, rollback defined, adversarial mitigations included, falsification tests included, blocking implementation unknowns absent. Generic end-of-pipeline MORE_RESEARCH is REMOVED as an outcome (§11). Frequent reopening of earlier stages is evidence of weak earlier gates and must be treated as a process defect. Implemented in a553fb1 (QUALIFY_OUTCOMES, _route_after_qualify). Dependency edges: QUALIFY→SYNTHESIS refinement; exceptional reopen→earliest affected stage only.
conductor commented 2026-09-09 01:31:58 +01:00
Author
Owner
Copy link
Copy source

#271 QUALIFY acceptance — REAL RUN on issue #161 (canonical RP8, commit 43ef8e4)

Verdict: PROCEED — pipeline COMPLETE.

Deterministic-first: 13/13 readiness checks passed BEFORE any MiMo call (design explicit, interfaces concrete, authority boundaries, data flow, sequence, invariants, tests, acceptance observable, failure modes, rollback, adversarial mitigations included, falsification tests included, blocking unknowns absent).

Semantic review (implementation-readiness ONLY, source-verified):

  • Interfaces unambiguous — every line anchor in the build instructions verified against actual source
  • recall() extension spec concrete: exact new signature given
  • Implementation sequence ordered by real dependencies (M1 mode field precedes the harness — the adversarial mitigation drives build order)
  • AC-1 through AC-12 each evaluable by automated test
  • Authority boundaries respected: harness is evaluation-only, leaf-positioned, registers via the existing memory_selector registry (M5 honored)
  • Rollback is real: deleting harness file + tests + harness-only registry registrations restores prior state

No reopen: no genuinely new blocking fact — REOPEN_EARLIER_STAGE not invoked (and correctly so; frequent reopens would indicate weak earlier gates, §10).

Narrowed semantics honored: no re-research, no re-litigation of the design, no generic MORE_RESEARCH. Malformed-verdict fail-safe and incomplete-reopen rejection proven in 9 unit tests.

Artifact persisted to master: research/issue-161/qualify.json + task_tree.json (status: COMPLETE, outcome: PROCEED).

Milestone: this is the first REAL issue to traverse the full corrected pipeline — ISSUE → INWARD → OUTWARD → APPLICABILITY → ADVERSARIAL(dispositioned) → SYNTHESIS → QUALIFY(PROCEED) — with every stage gate-passed on genuine evidence, INNOVATION skipped on merit, and targeted-refinement semantics proven structurally.

Literal handoff (§15) is now armed: accepted synthesis e584bbf895b3 (with carried mitigations + kill-tests embedded) is the verbatim implementation input for the MiMo Code build of retrieval_benchmark.py.

**#271 QUALIFY acceptance — REAL RUN on issue #161** (canonical RP8, commit 43ef8e4) **Verdict: PROCEED — pipeline COMPLETE.** **Deterministic-first:** 13/13 readiness checks passed BEFORE any MiMo call (design explicit, interfaces concrete, authority boundaries, data flow, sequence, invariants, tests, acceptance observable, failure modes, rollback, adversarial mitigations included, falsification tests included, blocking unknowns absent). **Semantic review (implementation-readiness ONLY, source-verified):** - Interfaces unambiguous — every line anchor in the build instructions verified against actual source - recall() extension spec concrete: exact new signature given - Implementation sequence ordered by real dependencies (M1 mode field precedes the harness — the adversarial mitigation drives build order) - AC-1 through AC-12 each evaluable by automated test - Authority boundaries respected: harness is evaluation-only, leaf-positioned, registers via the existing memory_selector registry (M5 honored) - Rollback is real: deleting harness file + tests + harness-only registry registrations restores prior state **No reopen:** no genuinely new blocking fact — REOPEN_EARLIER_STAGE not invoked (and correctly so; frequent reopens would indicate weak earlier gates, §10). **Narrowed semantics honored:** no re-research, no re-litigation of the design, no generic MORE_RESEARCH. Malformed-verdict fail-safe and incomplete-reopen rejection proven in 9 unit tests. Artifact persisted to master: research/issue-161/qualify.json + task_tree.json (status: COMPLETE, outcome: PROCEED). **Milestone:** this is the first REAL issue to traverse the full corrected pipeline — ISSUE → INWARD → OUTWARD → APPLICABILITY → ADVERSARIAL(dispositioned) → SYNTHESIS → QUALIFY(PROCEED) — with every stage gate-passed on genuine evidence, INNOVATION skipped on merit, and targeted-refinement semantics proven structurally. **Literal handoff (§15) is now armed:** accepted synthesis e584bbf895b3 (with carried mitigations + kill-tests embedded) is the verbatim implementation input for the MiMo Code build of retrieval_benchmark.py.
Sign in to join this conversation.
No Branch/Tag Specified
Labels
Clear labels
P0 — critical
Spine/core correctness or blocks other work — do first
 P1 — high
Material user value or real defect (money, bugs, next-stage dep)
 P2 — medium
Quality/consolidation improvement
 P3 — low
Measurement / benchmark / deferred policy
 blocked
Dependencies open
 bug
Confirmed defect or regression
 custodian
epic
Rune v1 course-correction issue graph
 feature
New capability or improvement
 governance
growth
identity
in-progress
Currently being worked on by Rune
 incomplete
Incomplete work — draft PR with ledger, awaiting resume
 infrastructure
life
memory
memory-atlas
Memory Research Atlas child domain
 needs-testing
Needs manual testing with dumbed-down steps before close
 perception
ready
Order satisfied - Rune may start
 release-drift
research
Research jobs — spikes, watchlists, evidence work
 rune
rune
 scaffolding
scaffolding
 spine,correction
spine,correction
 spine,deferred
spine,deferred
 spine,resolved
spine,resolved
 strangler,burn,endgame,deferred
strangler,burn,endgame,deferred
 strangler,burn,endgame,inventory,epic
strangler,burn,endgame,inventory,epic
 tui
tui
 tui,epic
tui,epic
 tui,openspec
tui,openspec
 v1-course-correction
Rune v1 course-correction issue graph
No labels bug
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
conductor
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: conductor/rune#271