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>
Two conflated issues: (1) the raw-URL 303 redirect is already followed
correctly by establish.py's urllib-based probe -- re-running
publish-check now passes; the stale roster fail was from the 2026-06-16
sweep, not a live issue. (2) config-atlas was never actually registered
with the production hub (absent from the 60 sources at
reuse.coulomb.social/v1/federated despite hub_registered: true in the
roster) -- corrected to false. Local publish pass is now 61/61; hub
registration itself is blocked on REUSE_SURFACE_TOKEN, which has no
routing catalog entry -- needs Bernd to provide or route it.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Prepared history/2026-07-06-t04-review-summary.md for the T05 human
review pass: full table of all 48 capability entries and 13 no-capability
markers generated from the actual committed files, plus flagged items
(the-custodian confidentiality scoping, vergabe-teilnahme low D1/C0,
markitect-main superseded framing, Forgejo-relevant tooling found along
the way).
Also fixed two pre-existing roster rows with stale empty
seed_capability_ids despite having real published entries: activity-core
and ops-warden. Not new drafts -- a bookkeeping gap predating this workplan.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
issue-core (migrated existing CAPABILITY-issue-tracking.yaml),
kaizen-agentic, key-cape, kontextual-engine, llm-connect, markitect-filter,
markitect-main (registered as superseded legacy platform), markitect-quarkdown,
markitect-tool, net-kingdom. All validate, no overlap with existing
federated capabilities. Committed locally in each sibling repo; push held
for T05 human review. Coverage 34/61 -> 44/61.
Also fixed a stale empty_scaffold_count >= 40 test threshold -- no longer
meaningful as the coverage campaign shrinks that number by design.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
artifact-store, can-you-assist, citation-engine, citation-evidence,
email-connect, guide-board, hub-core, info-tech-canon, infospace-bench,
inter-hub. Drafted directly (llm-connect not running locally; this
matches the workplan's no-invented-evidence principle better anyway).
All validate, no overlap with existing 24 federated capabilities.
Committed locally in each sibling repo; push held for T05 human review.
Coverage 24/61 -> 34/61. 27 has-capability repos remain for later cohorts.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Roster capability_status now has:11 / none:13 / pending:37. Coverage
24/61. NO_CAPABILITIES.md committed locally in all 13 sibling repos
(push held pending confirmation — touches 13 external default branches).
repo-seed's capability_count was stale (0, should be 1); corrected.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds capability_status: has|none|pending to every roster row (10 has /
51 pending currently), a coverage_ratio headline, and splits the gap
report's empty-scaffold section into unclassified vs explicitly-none.
Adds templates/NO_CAPABILITIES.template.md for the no-capability marker.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Add config-atlas as a federation source (capability index) and to the local
repo roster. Document surface.* as a distinct id namespace owned by
config-atlas (typed sibling of capability.*, not federated here) under a new
"Id namespace ownership" section. Raw URL currently 303 (required:false), same
publish block as state-hub/feature-control. Recompose federated.yaml.
Supports config-atlas ATLAS-WP-0002-T05.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Align agent files with on-disk workplan prefixes (infer from workplan ids)
- Set workplan domain to registered domain_slug; add topic_slug where applicable
- Repair frontmatter delimiter formatting; migrate legacy task status literals
- Regenerate AGENTS.md, CLAUDE.md, and .claude/rules from State Hub templates
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.
Mark 11 formerly-blocked repos publish pass after Gitea visibility fix.
Point whynot-control/design federation URLs at whynot/ org paths.
Remove activity-core owner stub from reuse-surface index (owner index
now fetchable). tegwick-control remains blocked (no HTTP-visible repo).
