llm-connect came up locally (mock provider, 127.0.0.1:8080), unblocking
this task.
reuse_surface/plan_check.py: build_rerank_prompt/request_rerank/apply_rerank,
reusing llm_bridge.execute_prompt/extract_json_object (same pattern as
maintain_llm.py's request_maintain_patches). New schema
schemas/plan-check-rerank.schema.json rejects malformed responses (missing
fields, non-JSON, invented candidate ids outside the input set) rather than
guessing.
Per design principle 3 (deterministic matches always rank first),
apply_rerank appends LLM-scored entries after the deterministic list
(kind: 'llm') instead of reordering it -- the trusted base result is
identical with or without the rerank pass. Graceful skip via a 'notes'
field when LLM_CONNECT_URL is unset or the response is malformed, mirroring
maintain.py's no_llm/skip pattern. New --llm-url/--no-llm CLI flags.
Also extended plan-check-result.schema.json for 'notes' and the
(pre-existing, previously unschema'd) 'filed_capability_request' field.
9 new pytest cases; 89 total pass. Live-verified against the running
llm-connect instance: it correctly rejected the mock provider's non-JSON
response and surfaced the skip note, with the deterministic verdict and
match order completely unaffected.
REUSE-WP-0018 is now fully done (T01-T06). Updated
docs/IntentScopeGapAnalysis.md priority 29 to Closed and the workplan's
own frontmatter status to finished.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Gate cleared: WP-0017 finished 2026-07-07, federated.yaml now carries 61
real capabilities (was 24). Dogfooded immediately -- plan-check against
'unified interface for issue tracking across Gitea GitHub GitLab' now
correctly returns extend/capability.infotech.issue-tracking instead of new.
Drafted the query-before-build convention in .claude/rules/workplan-convention.md
(gitignored, local -- propagation happens via the custodian rules-template
mechanism, not a reuse-surface git commit). Sent State Hub message to
custodian-agent proposing it be added to the shared template
(13cecd08-14e8-46d9-880c-65b8fef857cb).
Added reuse_check frontmatter to REUSE-WP-0018 and REUSE-WP-0019 themselves
as the reference implementation, both dogfooded for real against the live
federated index (both correctly verdict 'new').
REUSE-WP-0018 is now done except T03 (LLM semantic rerank), deferred
pending a running llm-connect instance. Updated IntentScopeGapAnalysis.md
priority 29 accordingly.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
T01: specs/PlanCheck.md design doc, plan-check-result.schema.json and
reuse-event.schema.json (the latter shared with WP-0019's reuse telemetry).
T02: reuse_surface/plan_check.py + 'reuse-surface plan-check' CLI command.
Deterministic token-Jaccard matching against registry/indexes/federated.yaml
(reuses overlaps.py's TOKEN_RE rather than a second scoring method), with
reuse/extend/new verdicts, markdown and --format json output, staleness
warning, and --record-outcome JSONL telemetry.
T04: reuse_surface/statehub_bridge.py bridges plan-check 'new' verdicts to
State Hub capability requests (--file-request) and surfaces open requests
with no matching capability (report gaps --check-capability-requests, opt-in
to stay offline-safe). Verified against the live local State Hub API;
status field (not catalog_entry_id presence) is the correct open/closed
signal, and the list endpoint needs a longer timeout than the health check
(~7s observed with 5 rows).
T06: docs (tools/README.md, RegistryFederation.md, SCOPE.md,
IntentScopeGapAnalysis.md priority 29) and an informational CI smoke step.
T03 (LLM rerank) not started -- llm-connect isn't running on this
workstation. T05 (ecosystem rollout) remains blocked: WP-0017 has drafted
entries for all 61 repos but they're still local-only pending its own T05
push/publish pass, so the federated index isn't yet worth rolling out
plan-check as ecosystem convention.
16 new tests, all mocked -- no network calls in the default test run.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
After Gitea UI transfer of tegwick/whynot repos to coulomb, align
workstation remotes and federation sources to coulomb/<slug>. Roster
sweep 60/60 pass; compose has 0 fetch warnings. Mark workplan finished.
Refresh federated index (20 capabilities, 0 duplicate warnings). Update
workplan, SCOPE, gap analysis, and tests for 59/60 publish pass. Mark
T04/T06 done; T01 remains on tegwick-control repo creation.
Note: production hub whynot URLs still need REUSE_SURFACE_TOKEN update.
Add hub sync and report cohorts CLI commands with pytest coverage, document
sibling index publish contract and hub hardening path, align INTENT layout,
raise external evidence on three registry entries, and close gap priorities
19-23 (priority 18 deferred on sibling index blocks).
Extend federation manifest schema for url sources with auth and TTL metadata.
Fetch remote capability indexes over HTTP(S), cache under
registry/federation/cache/, and fall back to stale cache on fetch failure.
Add --refresh flag, seven federation tests, and updated federation docs.
Register six new capabilities (12 total), add searchable catalog UI and graph
explorer, introduce pytest suite with CI fail-on-warnings, and close gap
analysis priorities 13 and 16. WP-0010 remains backlog for network federation.
Add federation manifest and schema, federation compose and graph CLI commands,
relation cycle/reference checks, federated index and Mermaid graph artifacts,
RegistryFederation guide, and CI validation updates.
Add Gitea CI workflow for registry validation, reuse-surface overlaps and
catalog commands, generated catalog artifacts, and documentation updates
closing gap analysis priorities 9-11.
Align INTENT.md with delivered layout, add CapabilityRegistryConcept guide,
extend schema with promotion_history, ship reuse-surface validate/query/export
CLI, register three more helix_forge capabilities, and refresh SCOPE and gap
analysis to reflect A3 tooling and D5/A3/C4/R2 self-assessment.
Document gaps between INTENT.md and SCOPE.md, refresh SCOPE with current
MVP capabilities, and seed the follow-up workplan for documentation
alignment, registry CLI tooling, and coverage growth.
