skill-security

AVE - Agentic Vulnerability Enumeration. The open behavioral classification standard for agentic AI components.

Open-source CLI scanner for agentic AI components such as skills, MCP servers, system prompts

See what your AI-agent skills can actually do. Vouch audits every SKILL.md on your machine, explains each skill's capabilities in plain English, and flags the risky ones (destructive commands, secret exfiltration, prompt injection). Local, deterministic — like npm audit, but for agent skills.

Find, vet, and install the right AI agent skill from a natural-language need.

Vetix — Automated scanning, identification, and assessment of SKILL security risks, based on Golang.

Static safety scanner for AI agent skills: detect prompt injection, curl-pipe-sh malware, credential theft before you install.

Audit AI skill safety before you ship. Static + semantic + adversarial audit pipeline for SKILL.md, Codex skill.yaml, Claude Code commands, and OpenCode. Free hosted tier on tarai.dev.

A curated AI coding skill and hook pack for safe local workflows.

Claude Code skills for keeping a Claude Code setup safe: review a third-party skill before installing it, lint and evaluate your own skill descriptions, audit the context budget, build least-privilege permissions, author and test hooks, find skill collisions and portability problems. Drives the author's gate, lock, simulator and hook tools.

Enhanced malicious Skill detection tool. Analyzes whether a target skill poses security threats to users who install it.

Built for the Gitlab AI Hackathon

A Codex and AI agent skill for privacy risk checks, PII redaction, output confirmation, skill installation review, audit logging, skill inventory, dangerous command protection, and external sharing confirmation.

Free Agent Skills specification and publish-readiness auditor by RankThread.

SQS (Skill Quality Suite): linter, validator, security scanner and eval harness for AI Agent Skills (SKILL.md) on Claude Code, Codex, Cursor and Gemini CLI - and it tells you which skill to fix or write next from your own sessions

Codex Skill：审计 Skill 和 AI agent 工具包的脚本、密钥、外部网络、权限边界与供应链风险

Static security scanner for AI skills, agents, prompts, and MCP instruction files

Runtime security intelligence platform for OpenClaw AI agents — skill auditing, policy enforcement, and threat detection

Community governance runtime template for OpenClaw inbound commands, task-package compilation, skill security, lifecycle gates and audit-ready execution boundaries.

Three portable Agent Skills for research, repository deployment, and pre-install safety review.

One-command pre-install scorecard for AI agent skills: security, permissions, token cost, footprint, and maintainability.