skill-audit

Write, validate, and review Agent SKILL.md's directly in Visual Studio Code. SKILL.md Inspector provides live diagnostics for every file named exactly SKILL.md, guided quick fixes, detailed reports, and workspace-wide checks for naming and scope conflicts.

Security scanner for Claude Code skills — detects malicious patterns, risk scoring, auto-install gating.

Audit any Claude Code SKILL.md file. Scored report with critical issues, triggering score, and exact fixes. Free skill-engineering tool by Anvil & Code.

Audit & optimize Hermes Agent skills — detect duplicates, estimate token waste, lifecycle management

Antigravity. Claude-code. 🇬🇧 Zero-dependency Node.js CLI to statically audit third-party AI Skills for malicious code patterns before local execution. | 🇪🇸 CLI Node.js sin dependencias para auditar estáticamente Skills de IA buscando código malicioso antes de ejecutarlos.

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.

Evidence-backed release audits for Agent Skills, covering discovery, behavioral uplift, safety, and package integrity. 为 Agent Skill 做有证据的上线审查，验证能否被正确触发、真实提升结果，并安全安装与复现。

🔍 Auditoria de segurança defensiva para Agent Skills (Claude, OpenClaw). Detecta payload em archives, ASCII smuggling, supply chain attacks e meta-prompt injection. Alinhado a OWASP ASI 2026 + AST01. Camada 1 da defesa em profundidade.

Local-first CLI to scan, audit, route, and test Codex Skills.

只读审计已安装/本地 Skill：把 SKILL.md、Skill 合集或专家包还原成可选、可调用、可验收的能力说明。支持 Codex / Claude Code / WorkBuddy 三平台一键安装。

Auditoria de qualidade para skills de agentes de IA - 10 criterios de excelencia

Skill auditor and health checker for AI agent skills: list every skill, detect duplicates and intent conflicts, score maturity, output keep/merge/refactor/remove advice. Skill 体检 / 冲突检测 / 成熟度评分

Security audit for AI agent skills. Detect malicious skills, prompt injection, data exfiltration, supply chain poisoning, two-stage payloads. 61 patterns aligned with OWASP Agentic AI Top 10. Works on Claude, ChatGPT, OpenAI, Gemini, Cursor, OpenClaw, ClawHub.

Audit your Claude Code skills for conflicts, overlaps, and redundancies. Splits auto-run from explicit-call, scans personal + plugin SKILL.md files, no API calls.

Check a Claude Code / Codex skill before you install it: which of your skills will it take requests from? Plus real trigger tests in your own setup.

Review Agent Skills with read-only package audits and clear English or Chinese HTML reports that explain what to improve.

OpenCode skill that creates, adapts, and audits agent skills from vague user requests. Generates SKILL.md files with spec, audit, and safety boundaries.

A decision-support framework for selecting the right approach in AI-assisted development workflows.

Read-only static safety auditor for AI skill folders and ZIP packages

Evidence-backed, resumable skill ecosystem auditing for the Codex app and CLI.