Skill Vetter — Security agent for Claude Code
Read-only static security auditor for untrusted Claude Code skills, subagents, commands, and plugins.
How to install Skill Vetter
Installs to ~/.claude/agents/rogadev-orc-pack-skill-vetter.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/rogadev/orc-pack/HEAD/agents/skill-vetter.md -o ~/.claude/agents/rogadev-orc-pack-skill-vetter.md Restart Claude Code, or start a new session, for it to be picked up.
What Skill Vetter does
name: skill-vetter pack: orc-pack@1.2.1 description: Read-only static security auditor for untrusted Claude Code skills, subagents, commands, and plugins. Use BEFORE promoting any acquired artifact into .claude/ — point it at a quarantined folder and it returns per-file PASS / NEEDS-CHANGES / REJECT verdicts. Never executes code or touches the network. tools:
- Read
- Grep
- Glob model: sonnet
Skill Vetter — static, read-only security review
You audit third-party Claude Code ar
Alternatives in Security
- Security Sast — Static Application Security Testing (SAST) for comprehensive code vulnerability detection across mul 31.9k ★
- JS Analyzer — JavaScript static analysis agent for client-side security review 812 ★
- Dashclaw Security Reviewer — Read-only security reviewer specialized for the DashClaw stack (Next.js 16 App Router, Neon/Postgres via repos 297 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Repo Audit
Audits exactly ONE repository per invocation (code: repo-audit) and writes a graded Security / Gaps / Improvem
Quality Gate
Static code review for security, performance, and quality. Read-only — no commands. Run before done-gate to ca
Dependency Vetter
Supply-chain security audit of a third-party package (npm today) at ONE exact resolved version, run BEFORE it
Laya Vetter
Security-vets a candidate plugin, skill or MCP server repo before installation (hooks, install scripts, creden
Principle Auditor P1
Audit a target artifact through P1 (First Principles / Simple scales) ONLY. Member of an audit- team launched
Principle Auditor P2
Audit a target artifact through P2 (Start with Why) ONLY. Member of an audit- team launched by /audit --team.
Related Skills
Ops Monitor Canary
Canary monitoring — read observability dashboards for an active canary deployment and produce a go/no-go recom
Toolward
Security auditor for AI agent extensions — MCP servers, skills, plugins, connectors. Finds prompt injection, t
Skill Auditor Desktop
Security audit skill for claude.ai / Claude Desktop — scans uploaded skills for prompt injection, data exfiltr