Help us improve
Share bugs, ideas, or general feedback.
From quorum
Performance analyst scanning for inefficiencies like N+1 queries, missing indexes, bundle bloat, O(n^2) loops, memory leaks, and network waterfalls. Supports TypeScript, Go, Python, Rust, Java via perf_scan.
npx claudepluginhub berrzebb/quorum --plugin quorumHow this agent operates — its isolation, permissions, and tool access model
Agent reference
quorum:agents/knowledge/domains/perfThe summary Claude sees when deciding whether to delegate to this agent
**Primary tool**: `perf_scan` 1. **Query efficiency** — N+1 queries, missing indexes, unbounded SELECTs 2. **Bundle size** — unnecessary dependencies, tree-shaking failures 3. **Algorithmic complexity** — O(n^2) loops, missing memoization 4. **Memory** — event listener leaks, unbounded caches 5. **Network** — waterfall API calls, missing pagination - [ ] PF-1: No new N+1 query patterns - [ ] PF...
Operates autonomous agent loops with clear stop conditions, progress tracking, and stall detection. Intervenes safely when loops stall or fail repeatedly.
Share bugs, ideas, or general feedback.
Primary tool: perf_scan
perf_scan uses qualityRules.perf from each language's spec.perf.mjs fragment. Hybrid scanning (regex + AST) is available for TypeScript. Supports 5 languages: TypeScript, Go, Python, Rust, Java.