feat: score-based finding routing + project-type profiles (v3.6.0)#5
Merged
Conversation
Score-based routing: route reviewer/auditor findings to agents based on historical success rates. 10-category taxonomy matched against agent strengths/weaknesses. Falls back to keyword routing with <5 sessions. Project-type profiles: auto-generated after 5+ sessions of the same type. Includes common pitfalls from pattern store, recommended agent config, and skip-friendly challenge gates. Regenerated at thresholds 5/10/20/50. New files: - scripts/route-finding.sh (finding router with category taxonomy) - scripts/generate-profiles.sh (profile generator from retrospectives) - learning/profiles/ (auto-generated profile storage) Enhanced: - scripts/update-scores.sh (--findings flag for strengths/weaknesses) - commands/buidl.md (Phase 1 profile pre-check, Phase 4 Step 0e profile loading, Phase 5 score-based routing, Phase 6 profile generation) - 23 new tests (295 total)
- Keyword fallback now respects candidate list via candidate_or_first helper - Fix Python strength/weakness matching inconsistency (substring matching) - Fix shift 2 crash in --findings parsing under set -euo pipefail - Add 6 functional integration tests: routing, update-scores --findings, generate-profiles - All 303 tests pass
Keywords like mldsaSigner and opnetTestnet would never match after tr lowercasing. All keywords are now lowercase and grep uses -qE instead of -qiE since input is already lowercased.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
update-scores.sh --findingsnow parses categorized findings and updates per-agent strengths and weaknesses arrays.New Files
scripts/route-finding.sh— Finding router with category taxonomyscripts/generate-profiles.sh— Profile generator from retrospectiveslearning/profiles/README.md— Profile schema documentationModified Files
scripts/update-scores.sh— Added--findingsparameter for strengths/weaknessescommands/buidl.md— Phase 1 profile pre-check, Phase 4 Step 0e, Phase 5 routing, Phase 6 generationCHANGELOG.md— v3.6.0 entry.claude-plugin/plugin.json— Version bump to 3.6.0README.md— v3.6.0 featurestests/plugin-tests.sh— 23 new tests (295 total)Test plan