docs(kd-oq): apply independent review fixes to open-question docs
Second-model read-only gap review of the KD-OQ decisions. All 5 decisions verified against source and confirmed to stand; corrected overstatements and two factual errors so the docs match the code:
- Prompt count 3 -> 4 (issue #2 Phase 5.3: daily-brief, search-synthesis, memory-session, knowledge-gaps) in roadmap + handover + KD-5a prompt. - KD-5 tool list now includes knowtation/consolidate (9 tools). - OQ1: detector classifies; a repo's active domain is still set via muse init --domain -> .muse/repo.json (not auto-activation). - OQ2: link_index/code_intel walk live vault files (not commit snapshots); only prime_context uses the commit graph. Flag that data/ is not auto-ignored (only .knowtation is) for KD-5a. - OQ4: interface + impl + registration helpers exist, but only the merge hook auto-registers at import (__init__.py); wiring the attestation provider / ICP push hook is KD-5a/5b work. - Added explicit KD-5a open contract items (search fan-out/ranking, consolidate return schema, attest runtime wiring, data/ ignore, detector-vs-repo.json activation) to the paste-ready prompt.
Verified via source: __init__.py registers only register_knowtation_hook; build_link_index(root) walks live files; stats.py reads data/ mtime only.
0 comments
muse hub commit comment sha256:5b35c11b2e5fb19ffdede036628383f89eca2ac56df708c8c8035d0f97eff545 --body "your comment"
No comments yet. Be the first to start the discussion.