Agent
Detects stale comments, docstring-signature mismatches, and contradicted NOTE/TODO/FIXME markers.
Risk-adaptive multi-agent PR review for Claude Code & Cursor — 7 specialized agents catch bugs, security flaws, and AI hallucinations in parallel
Agent
Detects stale comments, docstring-signature mismatches, and contradicted NOTE/TODO/FIXME markers.
Agent
Checks changed code against project conventions and coding style.
Agent
Detects logic errors, off-by-one bugs, null handling issues, and race conditions.
Agent
Detects when changes break callers, interfaces, or type contracts in other files.
Agent
Detects AI-generated code issues — non-existent APIs, wrong signatures, deprecated dependencies.
Agent
Uses git history to identify high-risk patterns in changed files.
Agent
Confirms or downgrades CRITICAL findings post-synthesis using deployment context; requires a cited mitigation for any downgrade.
Agent
Fast risk assessment of PR changes. Computes a 0-100 risk score from 6 weighted factors and recommends which review agents to dispatch.
Agent
Detects security vulnerabilities following OWASP Top 10 with data flow analysis.
Agent
Detects silent error-handling failures — empty catches, swallowed Promise rejections, optional-chaining-masked nullability, assertion-free tests.
Agent
Verifies acceptance criteria from REVIEW.md / .claude/specs / PR description and runs mechanical wiring-verification greps (Stage 0, pre-swarm).
Agent
Merges, deduplicates, filters, and synthesizes findings from all review agents into a coherent review.
Agent
Evaluates test coverage and test quality for changed code.