Internal process for the bug-clusterer agent. Defines the step-by-step procedure for parsing, classifying, redacting, scoring, and clustering bug candidates from raw X/Twitter posts. Not user-invocable — loaded by the agent via its `skills: ["bug-clustering"]` frontmatter property.
Step-by-step procedure for transforming raw XPost objects into structured, clustered bug candidates with PII redaction and reliability scoring.
For each XPost, produce a BugCandidate with all 33 fields using lib/parser.ts:
Before classification, run content-similarity deduplication using lib/dedupe.ts:
deduplicateCandidates() with parsed candidates and the candidate_dedup.hybrid_similarity_threshold from config/cluster-matching-thresholds.json (default 0.70)forward_ids) proceed to classification"{n} posts ({m} unique, {k} duplicate groups)"Run lib/classifier.ts on each candidate:
Run lib/redactor.ts on each candidate:
[REDACTED:type] tagsRun lib/reporter-scorer.ts on each candidate:
Match author against approved_accounts config:
Using lib/clusterer.ts and lib/signatures.ts:
lib/db.tsLoad evidence tier definitions for proper cluster evidence assessment:
!cat skills/x-bug-triage/references/evidence-policy.md
Load data model reference for BugCandidate fields and cluster schemas:
!cat skills/x-bug-triage/references/schemas.md