Private
Public Access
ed
84c0b4ecc4
Campaign: address the parent code_path_audit_20260607 Finding 1 (CRITICAL)
Metadata 4.01e22 effective codepaths via 3 SSDL techniques.
3 children, sequential, with budget gates:
1. metadata_nil_sentinel_20260624 (>= 10% drop): introduce
NIL_METADATA sentinel + migrate 6 nil-check functions.
2. metadata_generational_handle_20260624 (>= 20% drop,
BLOCKED_BY 1): wrap Metadata in (index, generation) handle;
collapse lifetime branches to 1 lookup + 1 cmp.
3. metadata_field_cache_20260624 (>= 30% drop, BLOCKED_BY 2):
MetadataFieldCache keyed by (handle.index, field_name);
123 string-keyed entry.get('key', default) sites become
cache lookups.
Each child has its own spec/plan/metadata/state. Budget gate
after each child: re-measure effective codepaths; if drop < threshold,
PAUSE the campaign and report to user.
End-of-campaign TRACK_COMPLETION captures the cumulative reduction
vs the 4.01e22 baseline. Deferred follow-up: apply the same
3 SSDL primitives to the 4 other dict[str, Any] aliases
(FileItem, CommsLogEntry, HistoryMessage, ToolDefinition, ToolCall).
16 files committed: 4 directories x 4 files each (spec, plan,
metadata, state).
3.6 KiB
3.6 KiB
Plan: metadata_generational_handle_20260624
3 tasks, 3 atomic commits. Same pattern as child 1.
Phase 1: Behavioral Test (1 task)
- Task 1.1: Write
tests/test_metadata_generational_handle.py.- WHERE: New file
tests/test_metadata_generational_handle.py - WHAT: 4 tests (per spec FR3)
- HOW: Use 1-space indentation
- SAFETY: Test imports
MetadataHandleandMetadataHandleRegistryfrom their production location - COMMIT:
test(metadata): behavioral test for generational handle (MetadataHandle + Registry) - GIT NOTE: 4 tests; will be RED until Phase 2 ships
- VERIFY:
uv run pytest tests/test_metadata_generational_handle.py -vshows 4/4 FAIL (expected)
- WHERE: New file
Phase 2: Implementation (1 task)
- Task 2.1: Add
MetadataHandle+MetadataHandleRegistry+ migrate lifetime-check code.- WHERE:
src/aggregate.py(or another existing module; no newsrc/<thing>.pyfiles) - WHAT:
- Add
@dataclass(frozen=True) class MetadataHandlewithindex: intandgeneration: int - Add
class MetadataHandleRegistrywithregister,lookup,bump_generationmethods - Migrate lifetime-check sites in
src/aggregate.py,src/ai_client.py, and any other production files
- Add
- HOW: Use
manual-slop_py_add_deffor the new types;manual-slop_edit_filefor the migration sites - SAFETY: Verify with
ast.parse; run the 4 tests + the child-1 tests + the 14 previously-failing tests - COMMIT:
feat(metadata): MetadataHandle + Registry + lifetime-check migrations - GIT NOTE: Lifetime checks now go through registry.lookup; generation mismatch returns NIL_METADATA from child 1
- VERIFY:
uv run pytest tests/test_metadata_generational_handle.py tests/test_metadata_nil_sentinel.py -vshows all PASS
- WHERE:
Phase 3: Verification + Budget Gate (1 task)
- Task 3.1: Run all 6 VCs; capture the budget gate measurement.
- WHERE: All audit gates + test suite + SSDL measurement
- WHAT:
- Run VC1-VC6
- Compute the new effective-codepaths number
- Compare to post-child-1 measurement; require drop ≥ 20%
- Write TRACK_COMPLETION report
- Update state.toml to
status = "completed", all 3 phases completed - Append to campaign_measurements_20260624.md
- Update conductor/tracks.md
- HOW: Same as child 1 phase 3
- SAFETY: Same as child 1
- COMMIT: 3 commits (state, TRACK_COMPLETION, tracks.md)
- GIT NOTE: Per workflow.md
- VERIFY: All 6 VCs pass; budget gate met; campaign unblocked for child 3
Commit Log (Expected)
test(metadata): behavioral test for generational handle(Task 1.1)feat(metadata): MetadataHandle + Registry + lifetime-check migrations(Task 2.1)conductor(state): metadata_generational_handle_20260624 SHIPPED(Task 3.1)docs(reports): TRACK_COMPLETION for metadata_generational_handle_20260624(Task 3.1)conductor(tracks): add metadata_generational_handle_20260624 row(Task 3.1)
Verification Commands
# VC1: types exist
grep -rn "class MetadataHandle\|class MetadataHandleRegistry" src/
# VC2: production uses handle
grep -rn "registry.lookup\|registry.register" src/
# VC3: tests pass
uv run pytest tests/test_metadata_generational_handle.py -v
# VC4: budget gate
uv run python -c "from src.code_path_audit_ssdl import compute_effective_codepaths; ...; print(compute_effective_codepaths(metadata_profile, 'src'))"
# VC5: full test suite
uv run python scripts/run_tests_batched.py
# VC6: 4 audit gates
uv run python scripts/audit_weak_types.py --strict
uv run python scripts/generate_type_registry.py --check
uv run python scripts/audit_main_thread_imports.py
uv run python scripts/audit_no_models_config_io.py