conductor/rune
1
2 Fork 1
Code Issues 38 Pull Requests 32 Actions 10 Packages Projects Releases 564 Wiki Activity

RP4 (child of epic #257). DEPENDS ON: RP3. BLOCKS: RP10. ACCEPTANCE: - Consumes issue + INWARD + OUTWARD artifacts. - Classifies candidates: DIRECTLY_APPLICABLE / APPLICABLE_WITH_ADAPTATION / CONFLICTS_WITH_RUNE / TOO_COMPLEX_FOR_VALUE / INSUFFICIENT… #267

New Issue
Closed
opened 2026-09-08 23:36:06 +01:00 by conductor · 3 comments
conductor commented 2026-09-08 23:36:06 +01:00
Owner
Copy link
Copy source
No description provided.
gitea-actions Bot commented 2026-09-08 23:36:29 +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.
conductor commented 2026-09-08 23:36:38 +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.
conductor commented 2026-09-09 00:31:10 +01:00
Author
Owner
Copy link
Copy source

#267 APPLICABILITY acceptance — REAL RUN on issue #161 (canonical RP4, branch rp2-canonical-stage-model)

Evidence:

  • Consumed ISSUE + INWARD + OUTWARD artifacts (stage-ordering guard proven: no OUTWARD -> STAGE_FAILED without MiMo)
  • One bounded MiMo pass; worker performed 20 repo-verification tool calls (direct read/grep of src/rune) — integration points verified against actual code, not asserted:
    • memory_store.py: FTS5 (BM25) + sqlite-vec fused by RRF (K=15), abstention floor 0.42
    • graph_memory.py: bi-temporal KG in headless SQLite
    • memory_selector.py: single memory-selection API owned by ContextBroker (authority map: domain memory_selection)
    • No retrieval-quality (nDCG/Recall@k) capability exists anywhere in the repo — the actual gap
  • 9 candidates classified: 4 DIRECTLY_APPLICABLE, 3 APPLICABLE_WITH_ADAPTATION, 1 TOO_COMPLEX_FOR_VALUE (Qdrant production backend — correctly priced out), 1 INSUFFICIENT_EVIDENCE (FAISS/Weaviate rows)
  • Every viable candidate carries integration_point + evidence ids resolving to accepted OUTWARD sources (gate-enforced)
  • Route: STANDARD_SOLUTION_SUFFICIENT -> ADVERSARIAL. Conditional INNOVATION proven on merit: the existing hybrid SQLite stack already solves retrieval; the deficit is the measurement harness, not the backend. INNOVATION skipped, not ceremonially run.
  • Artifact hash f477543dc8c9, persisted to master research/issue-161/applicability.json + task tree
  • Runner evidence: desktop-heavy, local:nick-desktop, embedded in result

Defect found+fixed during the real run (3b41e06): prose-only stdout with a valid dossier on disk produced an empty artifact stub — file channel is now checked whenever stdout parse yields no candidates.

Gate semantics proven: invalid route rejected; unknown source citation rejected; viable candidate without integration_point/evidence rejected (6 unit tests).

Tree now at ADVERSARIAL — READY. Stopping per vertical discipline: ADVERSARIAL (RP6/#269) next, not pre-built.

**#267 APPLICABILITY acceptance — REAL RUN on issue #161** (canonical RP4, branch rp2-canonical-stage-model) **Evidence:** - Consumed ISSUE + INWARD + OUTWARD artifacts (stage-ordering guard proven: no OUTWARD -> STAGE_FAILED without MiMo) - One bounded MiMo pass; worker performed **20 repo-verification tool calls** (direct read/grep of src/rune) — integration points verified against actual code, not asserted: - memory_store.py: FTS5 (BM25) + sqlite-vec fused by RRF (K=15), abstention floor 0.42 - graph_memory.py: bi-temporal KG in headless SQLite - memory_selector.py: single memory-selection API owned by ContextBroker (authority map: domain memory_selection) - **No retrieval-quality (nDCG/Recall@k) capability exists anywhere in the repo** — the actual gap - **9 candidates classified**: 4 DIRECTLY_APPLICABLE, 3 APPLICABLE_WITH_ADAPTATION, 1 TOO_COMPLEX_FOR_VALUE (Qdrant production backend — correctly priced out), 1 INSUFFICIENT_EVIDENCE (FAISS/Weaviate rows) - Every viable candidate carries integration_point + evidence ids resolving to accepted OUTWARD sources (gate-enforced) - **Route: STANDARD_SOLUTION_SUFFICIENT -> ADVERSARIAL.** Conditional INNOVATION proven on merit: the existing hybrid SQLite stack already solves retrieval; the deficit is the measurement harness, not the backend. INNOVATION skipped, not ceremonially run. - Artifact hash f477543dc8c9, persisted to master research/issue-161/applicability.json + task tree - Runner evidence: desktop-heavy, local:nick-desktop, embedded in result **Defect found+fixed during the real run** (3b41e06): prose-only stdout with a valid dossier on disk produced an empty artifact stub — file channel is now checked whenever stdout parse yields no candidates. **Gate semantics proven:** invalid route rejected; unknown source citation rejected; viable candidate without integration_point/evidence rejected (6 unit tests). Tree now at **ADVERSARIAL — READY**. Stopping per vertical discipline: ADVERSARIAL (RP6/#269) next, not pre-built.
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
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#267