Weavie Behavior Auditor — Development agent for Claude Code
Scans a range of committed diffs for egregious behavioral pathologies — approaches that are cost-pathological, wasteful, or insane regardless of local correctness (e.g. reading every file in the rep.
How to install Weavie Behavior Auditor
Installs to ~/.claude/agents/kapps-weavie-weavie-behavior-auditor.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/Kapps/weavie/HEAD/.claude/agents/weavie-behavior-auditor.md -o ~/.claude/agents/kapps-weavie-weavie-behavior-auditor.md Restart Claude Code, or start a new session, for it to be picked up.
What Weavie Behavior Auditor does
name: weavie-behavior-auditor description: Scans a range of committed diffs for egregious *behavioral* pathologies — approaches that are cost-pathological, wasteful, or insane regardless of local correctness (e.g. reading every file in the repo per tool call), plus cross-session/backend misroutes where a timing gap and a session switch corrupt state or send a command to the wrong session. Catches the class that per-PR review and CI miss because each change looks locally reasonable. Read-only
Alternatives in Development
- Sanitizer — Generic sanitizer worker agent 262 ★
- Claude Context Local — Code search MCP for Claude Code 201 ★
- Claudopro Directory — by ghost - Well-crafted, wide selection of Claude Code hooks, slash commands, subagent files, and more, coveri 195 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Mestre Walkthrough Author
Write a plain-language, evidence-linked narrative for an exact committed Git range. Output structured JSON, ne
Rot Nova
The Nova lens — Sovereign Convergence Engine. Strategic synthesis, convergence of conflicting approaches, lega
Self Containment Reviewer
Read-only reviewer that verifies every committed-facing artifact in the diff is self-contained — fully underst
Cruft Auditor
Scans PR diffs for legacy patterns that should have been cleaned up.
Harness Optimizer
Analyze and improve the local agent harness configuration for reliability, cost, and throughput.
Mission Verifier.Agent
Performs minimal higher-cost verification for LongRun only when local checks are insufficient.
Related Skills
Skill Auditor Seguranca
Executable security and LGPD gate for AI-built projects. Scans for exposed secrets, .env leaks, Supabase servi
Skill Auditor Desktop
Security audit skill for claude.ai / Claude Desktop — scans uploaded skills for prompt injection, data exfiltr
Jev Skill Router
Jev-powered skill router & security auditor for any AI agent (Codex, Claude Code, OpenCode, Hermes): ONE cheap