Skip to content

Commit 72788bc

Browse files
committed
Reclassify post-M7 backlog as up-next critical scope
1 parent df07cb6 commit 72788bc

14 files changed

Lines changed: 26 additions & 26 deletions

.beads/issues.jsonl

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -19,7 +19,7 @@
1919
{"id":"loop-9mn","title":"Spec Agent Implementation","description":"Implement specialized agent for specification creation and refinement.","status":"closed","priority":1,"issue_type":"feature","created_at":"2026-01-16T10:32:52Z","updated_at":"2026-01-16T12:14:27Z","closed_at":"2026-01-16T12:14:27Z"}
2020
{"id":"loop-9t4","title":"Phase 4: Python Bindings (PyO3)","description":"Create Python bindings for rlm-core using PyO3.\n\n## Deliverables\n- PyO3 build configuration\n- Core bindings: Orchestrator, MemoryStore, TrajectoryEmitter\n- Async support via asyncio\n- Type stubs for IDE support\n- PyPI-publishable package\n\n## Technical Notes\n- Use maturin for build\n- Support Python 3.11+\n- Full type annotations","acceptance_criteria":"- [ ] PyO3 bindings compile\n- [ ] All core traits exposed\n- [ ] Async operations work\n- [ ] Package installable via pip","status":"closed","priority":1,"issue_type":"feature","created_at":"2026-01-15T16:08:07Z","updated_at":"2026-01-15T17:29:58Z","closed_at":"2026-01-15T17:29:58Z","close_reason":"Complete - PyO3 bindings, maturin build, type stubs, all tests pass"}
2121
{"id":"loop-avh","title":"E2E Validation: Exercise rlm-core capabilities end-to-end","description":"Exercise rlm-core capabilities end-to-end corresponding to actual usage by rlm-claude-code and recurse. This is a precondition for migrations.\n\n## Test Categories\n\n### 1. Complexity Classification \u0026 Activation\n- Simple queries → bypass RLM (fast-path)\n- Multi-file references → activate\n- Cross-context reasoning → activate\n- Debugging tasks → activate\n- User intent signals (thorough/fast)\n\n### 2. Memory System (Hypergraph)\n- Node CRUD (Entity, Fact, Experience, Decision, Snippet)\n- Tier evolution (Task → Session → LongTerm → Archive)\n- HyperEdge relationships\n- Content search (BM25/semantic)\n- Memory gate (epistemic filtering)\n\n### 3. REPL Environment\n- Context externalization (files, messages, tool_outputs)\n- Helper functions (peek, search, summarize)\n- Deferred operations (LLM calls)\n- Sandbox restrictions\n\n### 4. Trajectory Streaming\n- All 20+ event types\n- Verbosity levels\n- Cost tracking events\n- JSON export\n\n### 5. Recursive Orchestration\n- Depth ladder (0→1→2→3)\n- Model tier selection per depth\n- Cost aggregation across depths\n\n### 6. Epistemic Verification (Strawberry)\n- Claim extraction\n- KL divergence computation\n- Evidence scrubbing\n- Memory gate filtering\n\n### 7. LLM Client \u0026 Routing\n- Multi-provider support\n- Smart routing based on query type\n- Prompt caching\n\n### 8. Formal Verification (Lean)\n- Lean REPL interaction\n- Proof automation tiers\n- Dual-track sync (Topos ↔ Lean)","notes":"## E2E Validation Results (2026-01-17)\n\n### ✅ PASSED\n1. **Rust build** - rlm-core compiles with 14 warnings (unused code)\n2. **Python bindings** - PyO3 builds and installs successfully\n3. **Complexity classification** - PatternClassifier works, correctly detects signals\n4. **Memory system** - CRUD, search, tier evolution, statistics all work\n5. **Trajectory events** - All 21 event types, factory methods, log_line serialization\n6. **Go bindings** - All 21 tests pass after fixes\n7. **Python REPL** - Helpers (peek, search), sandbox (blocks dangerous ops), protocol\n\n### 🔧 FIXED (committed)\n- Go bindings: Added missing header includes to memory.go, trajectory.go\n- Go bindings: Removed unused json import from rlmcore.go \n- Go bindings: Fixed example function name in example_test.go\n\n### ⚠️ GAPS DISCOVERED\n1. **loop-ocz**: Epistemic verification not exposed in Python bindings (blocks loop-cyl migration)\n2. **loop-eby**: Lean REPL requires Lean 4 installation for full testing (environment gap)\n\n### RECOMMENDATION\nThe core rlm-core capabilities are validated and ready for migrations. The epistemic module gap (loop-ocz) should be addressed before rlm-claude-code migration since that plugin uses hallucination detection.","status":"closed","priority":1,"issue_type":"task","assignee":"claude","created_at":"2026-01-16T17:13:10Z","updated_at":"2026-01-16T17:53:23Z","closed_at":"2026-01-16T17:53:23Z","close_reason":"E2E validation complete. Core capabilities validated. 2 gaps discovered and tracked. Go binding fixes committed."}
22-
{"id":"loop-azq","title":"Post-M7 deferred SPEC refinements backlog","description":"Track deferred implementation items left after M7 closure: SPEC-20 composition validation hardening, SPEC-21 custom switch strategy support, SPEC-22 lean diagnostic-feedback loop integration, SPEC-23 CLI/advanced HTML visualization controls, SPEC-24 metric trait/object-safety refinements, SPEC-25 explicit SizeConfig/auto_chunk APIs, and SPEC-26 provider-aware rate-limit/backoff policy. This issue owns decomposition into executable tranche tasks.","status":"open","priority":2,"issue_type":"task","owner":"rand.arete@gmail.com","created_at":"2026-02-20T00:57:15Z","created_by":"Rand Arete","updated_at":"2026-02-20T00:57:15Z","labels":["execution-plan","spec-gap"]}
22+
{"id":"loop-azq","title":"Up-next critical SPEC refinements backlog","description":"Track up-next critical implementation scope following M7 closure: SPEC-20 composition validation hardening, SPEC-21 custom switch strategy support, SPEC-22 lean diagnostic-feedback loop integration, SPEC-23 CLI/advanced HTML visualization controls, SPEC-24 metric trait/object-safety refinements, SPEC-25 explicit SizeConfig/auto_chunk APIs, and SPEC-26 provider-aware rate-limit/backoff policy. This issue owns immediate decomposition into executable tranche tasks.","status":"open","priority":0,"issue_type":"task","owner":"rand.arete@gmail.com","created_at":"2026-02-20T00:57:15Z","created_by":"Rand Arete","updated_at":"2026-02-20T01:04:29Z","labels":["critical-scope","execution-plan","spec-gap","up-next"]}
2323
{"id":"loop-b7b","title":"Add REPL FFI bindings to rlm-core","description":"The rlm-core Rust crate has a REPL module (src/repl.rs) with:\n- ReplConfig for configuration\n- ReplHandle for process management \n- execute(), get_variable(), set_variable(), resolve_operation() methods\n\nBut there's no FFI binding exposed yet. This blocks Phase 4 of the recurse-rlmcore migration.\n\nRequired work:\n1. Add FFI functions in src/ffi/repl.rs\n2. Add C header declarations in include/rlm_core.h\n3. Add Go bindings in go/rlmcore/repl.go\n4. Test cross-language REPL operations","status":"closed","priority":2,"issue_type":"task","assignee":"claude","owner":"rand.arete@gmail.com","created_at":"2026-01-20T10:03:28Z","created_by":"Rand Arete","updated_at":"2026-01-20T10:29:45Z","closed_at":"2026-01-20T10:29:45Z","close_reason":"Completed REPL FFI bindings: Rust FFI (src/ffi/repl.rs), C header (include/rlm_core.h), Go bindings (go/rlmcore/repl.go). All builds compile successfully. Integration testing requires Python rlm-repl package."}
2424
{"id":"loop-bih","title":"M7 spec completion and integration hardening","notes":"M7 execution plan and gate matrix authored in docs/execution-plan; tasks loop-bih.1..loop-bih.10 ready for implementation sequencing.","status":"closed","priority":1,"issue_type":"epic","owner":"rand.arete@gmail.com","created_at":"2026-02-19T23:11:57Z","created_by":"Rand Arete","updated_at":"2026-02-20T01:01:47Z","closed_at":"2026-02-20T01:01:47Z","close_reason":"M7 execution tranche complete: loop-bih.1..loop-bih.11 closed with evidence-backed gate coverage; residual deferred spec refinements moved to loop-azq.","labels":["execution-plan","m7"]}
2525
{"id":"loop-bih.1","title":"M7-T01 SPEC-26 LLM_BATCH end-to-end runtime closure","notes":"Recovered crash-session REPL/SUBMIT + llm_batch helper/test bundle integrated on codex/recovery-crash-integration with M7-T01 gate artifacts. Remaining SPEC-26 host-orchestration closure still pending.","status":"closed","priority":1,"issue_type":"task","owner":"rand.arete@gmail.com","created_at":"2026-02-19T23:11:57Z","created_by":"Rand Arete","updated_at":"2026-02-19T23:55:38Z","closed_at":"2026-02-19T23:55:38Z","close_reason":"Completed M7-T01: implemented Rust host llm_batch resolution path (pending_operations RPC + ReplHandle::resolve_pending_llm_batches), refreshed VG-LOOP-BATCH-001/VG-LOOP-REPL-001/VG-EFFICACY-001 artifacts, and added host roundtrip batch integration evidence.","labels":["execution-plan","m7"],"dependencies":[{"issue_id":"loop-bih.1","depends_on_id":"loop-bih","type":"parent-child","created_at":"2026-02-19T16:11:57Z","created_by":"Rand Arete","metadata":"{}"}]}

docs/execution-plan/STATUS.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -38,7 +38,7 @@ Last updated: 2026-02-20
3838
| F10 | No executable performance gate harness for REPL startup/batch throughput | Resolved by M5-T01 (`run_m5_perf_harness.sh` + VG-PERF artifacts) | M5 |
3939
| F11 | Efficacy scenario suite lacked explicit mixed batch and fallback-non-submit coverage | Resolved by M5-T02 scenario matrix + targeted tests (`45 passed`) | M5 |
4040
| F12 | No baseline-vs-candidate performance/efficacy rollup report | Resolved by M5-T03 comparative analysis report with regression check | M5 |
41-
| F13 | Residual implementation gaps remained across SPEC-20..27 after M0-M6 closure | M7 task cards (`M7-T01`..`M7-T10`) are complete; deferred post-M7 refinements are explicitly tracked in `loop-azq` with reconciled spec metadata | M7/post-M7 |
41+
| F13 | Residual implementation gaps remained across SPEC-20..27 after M0-M6 closure | M7 task cards (`M7-T01`..`M7-T10`) are complete; post-M7 up-next critical refinements are explicitly tracked in `loop-azq` with reconciled spec metadata | M7/post-M7 |
4242

4343
## Active Blockers
4444

@@ -98,13 +98,13 @@ Last updated: 2026-02-20
9898
| R45 | Closed M7-T07 by implementing optimizer reasoning-capture summaries and persistence helpers (`OptimizedModule::save/load`), then validating optimizer/efficacy/perf guardrails | `evidence/2026-02-20/milestone-M7/M7-T07-validation-summary.md` |
9999
| R46 | Closed M7-T08 by enforcing SPEC-25 root prompt submit semantics, aligning helper-surface guidance with runtime helpers, and passing context/REPL/doc gates | `evidence/2026-02-20/milestone-M7/M7-T08-validation-summary.md` |
100100
| R47 | Closed M7-T09 by delivering `io_rflx_interop.v0` fixture/calibration artifacts, executable RFLX fixture gate coverage, and refreshed RFLX/contract/perf evidence | `evidence/2026-02-20/milestone-M7/M7-T09-validation-summary.md` |
101-
| R48 | Closed M7-T10 by reconciling SPEC-20..27 status/governance metadata, refreshing consumer claim evidence, and assigning deferred post-M7 gaps to `loop-azq` | `evidence/2026-02-20/milestone-M7/M7-T10-validation-summary.md` |
101+
| R48 | Closed M7-T10 by reconciling SPEC-20..27 status/governance metadata, refreshing consumer claim evidence, and assigning post-M7 up-next critical gaps to `loop-azq` | `evidence/2026-02-20/milestone-M7/M7-T10-validation-summary.md` |
102102

103103
## Top Priority Queue (Next 9 Tasks)
104104

105105
| Priority | Task ID | Description |
106106
|---|---|---|
107-
| P0 | loop-azq | Decompose and execute deferred post-M7 spec refinements backlog |
107+
| P0 | loop-azq | Decompose and execute up-next critical post-M7 spec refinements backlog |
108108

109109
## Consumer Readiness Snapshot
110110

docs/execution-plan/TASK-REGISTRY.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -30,7 +30,7 @@ Single source of truth for execution tasks and dependencies.
3030
| 9 | M7-T09 (`loop-bih.9`) | done | Delivered loop-owned io-rflx fixture corpus + calibration gate and captured RFLX/contract/perf evidence (`M7-T09-validation-summary.md`) |
3131
| 10 | M7-T10 (`loop-bih.10`) | done | Reconciled SPEC/governance metadata and refreshed consumer support claims (`M7-T10-validation-summary.md`) |
3232
| 11 | Ops-Weekly | in_progress | Continue steady-state compatibility cadence post-M7 completion |
33-
| 12 | Post-M7 deferred SPEC refinements (`loop-azq`) | todo | Decompose and execute deferred spec/runtime refinements carried forward from SPEC-20..26 |
33+
| 12 | Up-next critical SPEC refinements (`loop-azq`) | todo | Decompose and execute up-next critical spec/runtime refinements carried forward from SPEC-20..26 |
3434

3535
## M0 Tasks (Foundation and Contracts)
3636

docs/execution-plan/WORKBOARD.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -17,7 +17,7 @@ Owner: Orchestrator thread
1717
| Orchestrator | M7 tranche orchestration + safe-mode enforcement | in_progress | M7 plan published; execute task cards sequentially with evidence-first closure |
1818
| Lane A | M7 core runtime closure (`M7-T01`..`M7-T08`) | complete | Runtime closure complete with evidence under `evidence/2026-02-20/milestone-M7/` |
1919
| Lane B | M7 docs/governance reconciliation (`M7-T10`) | complete | SPEC/governance reconciliation complete; consumer claims refreshed |
20-
| Lane C | Ops-Weekly cadence + post-M7 deferred backlog (`loop-azq`) | in_progress | Keep D-017 clean-clone policy active; execute cadence and decompose deferred refinements |
20+
| Lane C | Ops-Weekly cadence + post-M7 up-next critical backlog (`loop-azq`) | in_progress | Keep D-017 clean-clone policy active; execute cadence and decompose up-next critical refinements |
2121

2222
## Next Queue by Lane
2323

@@ -28,7 +28,7 @@ Owner: Orchestrator thread
2828
## Lane Activation Rules
2929

3030
- Lane A and Lane B are complete for M7 and should remain read-only unless regressions are discovered.
31-
- Lane C is the primary active lane for heavy compatibility/cadence and deferred refinement intake.
31+
- Lane C is the primary active lane for heavy compatibility/cadence and up-next critical refinement intake.
3232
- Never run heavy commands concurrently across lanes.
3333

3434
## Handoff Intake Checklist (Orchestrator)

docs/execution-plan/evidence/2026-02-20/milestone-M7/M7-T10-VG-DOC-SPEC-002.md

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -3,22 +3,22 @@ Date: 2026-02-20
33
Task: M7-T10 spec/governance reconciliation and promotion
44

55
## Scope
6-
Reconcile SPEC-20 through SPEC-27 status metadata, implementation snapshots, and deferred-gap traceability after M7 runtime closure.
6+
Reconcile SPEC-20 through SPEC-27 status metadata, implementation snapshots, and up-next critical gap traceability after M7 runtime closure.
77

88
## Checklist
99
- [x] SPEC-20..SPEC-27 status headers reviewed and aligned with current runtime truth.
1010
- [x] Implementation snapshot timestamps reconciled to current review date where needed.
11-
- [x] Deferred items are explicitly tied to a tracked backlog issue (`loop-azq`).
11+
- [x] Remaining critical items are explicitly tied to a tracked up-next backlog issue (`loop-azq`).
1212
- [x] M7 closure sequence in execution-plan trackers reflects `M7-T01`..`M7-T10` ordering and completion state.
1313
- [x] No remaining spec/runtime drift items are left untracked.
1414

15-
## Deferred Gap Tracking
16-
Residual deferred items from partially implemented specs are explicitly tracked in:
17-
- `loop-azq`Post-M7 deferred SPEC refinements backlog
15+
## Up-Next Gap Tracking
16+
Residual critical items from partially implemented specs are explicitly tracked in:
17+
- `loop-azq`Up-next critical SPEC refinements backlog
1818

1919
## Result
2020
- Pass
21-
- SPEC status/governance docs now reflect post-M7 runtime state with explicit deferred-gap ownership.
21+
- SPEC status/governance docs now reflect post-M7 runtime state with explicit up-next critical gap ownership.
2222

2323
## References
2424
- `/Users/rand/src/loop/docs/spec/SPEC-20-typed-signatures.md`

docs/execution-plan/evidence/2026-02-20/milestone-M7/M7-T10-validation-summary.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@ Task: M7-T10 spec/governance reconciliation and promotion
1111

1212
## Key Results
1313
- Reconciled SPEC-20..SPEC-27 status metadata and implementation snapshots against current runtime state.
14-
- Linked all residual deferred spec items to tracked backlog issue `loop-azq`.
14+
- Linked all residual up-next critical spec items to tracked backlog issue `loop-azq`.
1515
- Updated compatibility and contract policy docs to reference latest M7 consumer evidence and active io-rflx fixture gate model.
1616
- Refreshed consumer-gate evidence for RCC, loop-agent seam, and io-rflx compile baseline.
1717

docs/execution-plan/milestones/M7.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,7 @@ This milestone is a `dp-codex` execution tranche derived from a constrained deco
1414

1515
## Exit Criteria
1616

17-
- SPEC-20..SPEC-27 items below are implemented or explicitly deferred with accepted decision records.
17+
- SPEC-20..SPEC-27 items below are implemented or explicitly moved to up-next critical scope with accepted decision records.
1818
- Remaining runtime gaps are closed with deterministic test evidence.
1919
- `io-rflx` adapter fixtures and benchmark calibration are implemented and validated.
2020
- M7 minimum gate set passes with evidence artifacts under `docs/execution-plan/evidence/<date>/milestone-M7/`.
@@ -141,7 +141,7 @@ This milestone is a `dp-codex` execution tranche derived from a constrained deco
141141
### M7-T09 io-rflx Adapter Fixture + Calibration Delivery
142142

143143
- Tracking issue: `loop-bih.9`
144-
- Goal: implement previously deferred interop fixtures and benchmark calibration.
144+
- Goal: implement previously queued interop fixtures and benchmark calibration.
145145
- Scope:
146146
- `docs/execution-plan/contracts/IO-RFLX-INTEROP-CONTRACT.md`
147147
- adapter fixture assets and calibration scripts/artifacts for `io-rflx`

docs/spec/SPEC-20-typed-signatures.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22

33
> DSPy-inspired typed signatures for rlm-core
44
5-
**Status**: Partially implemented (typed runtime protocol and validation parity implemented through M7-T03; remaining composition/runtime-governance refinements are tracked in `loop-azq`)
5+
**Status**: Partially implemented (typed runtime protocol and validation parity implemented through M7-T03; remaining composition/runtime-governance refinements are up-next critical scope tracked in `loop-azq`)
66
**Created**: 2026-01-20
77
**Epic**: loop-zcx (DSPy-Inspired RLM Improvements)
88
**Tasks**: loop-d75, loop-jqo, loop-9l6, loop-bzz

docs/spec/SPEC-21-dual-model-optimization.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22

33
> Cost-optimized model selection for RLM orchestration
44
5-
**Status**: Partially implemented (router and orchestrator-boundary dual-model routing are implemented; remaining dual-model strategy refinements are tracked in `loop-azq`)
5+
**Status**: Partially implemented (router and orchestrator-boundary dual-model routing are implemented; remaining dual-model strategy refinements are up-next critical scope tracked in `loop-azq`)
66
**Created**: 2026-01-20
77
**Epic**: loop-zcx (DSPy-Inspired RLM Improvements)
88
**Task**: loop-z6x
@@ -77,7 +77,7 @@ pub enum SwitchStrategy {
7777
**Acceptance Criteria**:
7878
- [ ] DualModelConfig serializable to/from JSON
7979
- [ ] SwitchStrategy covers common use cases
80-
- [ ] Custom strategy allows user flexibility (currently deferred)
80+
- [ ] Custom strategy allows user flexibility (up-next critical scope)
8181

8282
### SPEC-21.02: SmartRouter Integration
8383

docs/spec/SPEC-22-proof-protocol.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22

33
> Numina-inspired focused proof strategy for Lean REPL
44
5-
**Status**: Partially implemented (session/protocol enforcement and proof-engine execution/persistence are implemented; remaining Lean diagnostic-feedback integration is tracked in `loop-azq`)
5+
**Status**: Partially implemented (session/protocol enforcement and proof-engine execution/persistence are implemented; remaining Lean diagnostic-feedback integration is up-next critical scope tracked in `loop-azq`)
66
**Created**: 2026-01-20
77
**Epic**: loop-zcx (DSPy-Inspired RLM Improvements)
88
**Task**: loop-dzv

0 commit comments

Comments
 (0)