Repository navigation
fix(maproom): correct vector-search error behavior and repo-name/shared-index doc fixes (plugin-critical-fixes) - #66
danielbushman wants to merge 3 commits into
Conversation
…e (plugin-critical-fixes) - A1/A2: Replace --repo crewchief/crewchief-specs with manifoldlogic/crewchief and specs in multi-repo-guide.md and maproom-repos.yaml template; add owner/repo naming note with fuzzy-match explanation per R1 - A3: Replace First-Time Setup section with Shared Index guidance; demote db migrate/scan to clearly-labeled admin appendix per R2 - A4: Add FTS-first callout at Choosing Search Type table noting 0% embeddings; vector-search returns empty until embeddings are generated per R3 - R4: Bump plugin.json 0.10.0 → 0.10.1 (PATCH, doc fix) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…t empty results (plugin-critical-fixes) Live measurement shows maproom vector-search exits with code 2 and embedding_provider error when embeddings are absent, not empty results. Update SKILL.md lines 30, 46, 280 to reflect accurate behavior. Also update illustrative chunk counts in multi-repo-guide.md to match live index (29,764 / 28,799) from stale values (24,333 / 1,200). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
Warning Review limit reached
Next review available in: 53 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
WalkthroughMaproom search guidance now targets a populated shared index, uses canonical repository identifiers, updates multi-repository examples and setup commands, expands troubleshooting instructions, and bumps the plugin version to 0.10.1. ChangesMaproom search guidance
Estimated code review effort: 2 (Simple) | ~10 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@plugins/maproom/skills/maproom-search/SKILL.md`:
- Around line 15-46: The search guidance is inconsistent: the “Choosing Search
Type” table still sends conceptual queries to maproom vector-search even though
this section says embeddings are 0% and maproom search is the only usable path
right now. Update the table in SKILL.md so it matches the FTS-first guidance by
routing all current search use cases to maproom search, and keep maproom
vector-search clearly marked as unavailable until embeddings exist.
In `@plugins/maproom/skills/maproom-search/templates/maproom-repos.yaml`:
- Around line 29-34: The repo naming note in maproom-repos.yaml is too absolute
because it states repo keys are always owner/repo, but the same mapping includes
specs as a non-owner/repo key. Update the comment near the repo-key guidance to
either explicitly call out specs as an exception or rephrase it to say “most
repo keys” are owner/repo, using the existing repo-key note text as the target
for the change.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: bcae393a-eb7b-4f63-8cb3-5eb91c5c5025
📒 Files selected for processing (4)
plugins/maproom/.claude-plugin/plugin.jsonplugins/maproom/skills/maproom-search/SKILL.mdplugins/maproom/skills/maproom-search/references/multi-repo-guide.mdplugins/maproom/skills/maproom-search/templates/maproom-repos.yaml
…ten repo-key note - SKILL.md lines 30/46/281: replace full `maproom search` / `maproom vector-search` prose references with subcommand-style wording so verify-docs.sh Check 3 passes (bare commands are only allowed inside the Output Formats section). - SKILL.md search-type table: conceptual/exploring rows now route to FTS now and vector-search only when embeddings available, consistent with the FTS-first callout that says vector-search exits code 2 in this environment (CodeRabbit SKILL.md:46). - maproom-repos.yaml naming note: soften "always owner/repo" to "typically origin-derived owner/repo, except explicitly-pinned names like specs" (CodeRabbit maproom-repos.yaml:34). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Summary
What Changed
Prior commit (afae91a) — P1 and P2 fixes
--repoexamples updated to real names (manifoldlogic/crewchief,specs) verified against livemaproom status --jsondb migrate/scandemoted from first-time setup to a clearly-labeled admin appendix--all-reposreferencesThis commit (37a469d) — adversarial review MUST-FIX
SKILL.md lines 30, 46, 280 — Three locations incorrectly described
maproom vector-searchas "returning empty results" or "completing but returning empty" when embeddings are absent. Live measurement (3 runs, exit code verified):Vector-search exits with code 2 — it does not produce an empty result set. Agents relying on empty-result fallback logic would instead receive a hard error. All three locations updated to state the actual behavior.
multi-repo-guide.md line 361 — Illustrative chunk counts updated from stale values (24,333 / 1,200) to live values (29,764 / 28,799) verified against
maproom status.Spec Reference
_SPECS/crewchief/research/maproom-eco-plugin-critical-fixes-spec.md— Wave 1, chunkplugin-critical-fixesValidation Evidence
Review
Adversarial review findings (from JSON input):
Non-Goals
Per spec: glossary/troubleshooting/embedding-doc banners (Wave 2
plugin-backend-docs-refresh); researcher-agent changes (Wave 2); cross-repo guidance (Wave 3).🤖 Generated with Claude Code
Summary by CodeRabbit
Documentation
Chores
0.10.1.