[Pepps233]

Summary

Persist, consolidate, qualify, and prioritize evidence-grounded skill candidates across completed evolution jobs through an event-backed registry and retryable background worker.

Reasoning

Individual evolution jobs produce immutable candidates, but repeated evidence needs a durable workspace-scoped lifecycle before synthesis or activation can be considered. Consolidating compatible candidates preserves proposal revisions and counterevidence while allowing qualification to depend on independent task recurrence.

Registry processing runs separately from the main evolution pipeline so filesystem persistence, optimistic revision conflicts, retries, work checkpoints, and background-priority coordination do not block candidate generation or make candidates executable.

Changes

• Add an event-backed skill-candidate registry with immutable proposal revisions and workspace-scoped grouping.
• Consolidate compatible candidates and preserve supporting and contradicting evidence across jobs.
• Qualify groups from independent task recurrence and calculate deterministic evidence-priority scores.
• Persist registry events through the filesystem store with revision conflict detection, atomic writes, and corruption checks.
• Add checkpointed worker processing with bounded retries, retry backoff, idempotency, and terminal failure history.
• Run registry work through the background-priority gate and wire worker startup and shutdown into the coding-agent runtime.
• Document persisted registry state in evolution recovery guidance.

User impact

Evidence-grounded candidates are durably grouped and scored across jobs without synthesizing, evaluating, or activating executable skills. Contradicting evidence can quarantine candidate groups, while insufficient recurrence keeps them proposed for later evidence.

Stack

Position: 4 of 12
Base: #137
Next: #139

1. feat(coding-agent): activate versioned evolved skills #134
2. feat(agent): preserve target-neutral reflection evidence #136
3. feat(agent): curate and checkpoint skill candidates #137
4. feat(agent): persist and process skill candidate registry #138 (this PR)
5. feat(agent): synthesize outcome-aware skill drafts #139
6. feat(agent): validate evolved skill drafts #140
7. feat(agent): measure procedural information loss #147
8. feat(agent): evaluate evolved skill candidates #148
9. feat(agent): activate evolved skill candidates #149
10. feat(agent): observe evolved skill lifecycle outcomes #150
11. feat(agent): define evolved skill memory handling #151
12. feat(coding-agent): add evolved skill operator controls #152

Testing

• cd packages/evolution && node ../../node_modules/vitest/dist/cli.js --run test/skill-candidate-registry.test.ts test/skill-candidate-worker.test.ts test/filesystem-store.test.ts
• cd packages/coding-agent && node ../../node_modules/vitest/dist/cli.js --run test/evolution-runtime.test.ts

This pull request is AI-generated by Diffex.