skill-scanner

A full-stack AI Red Teaming platform securing AI ecosystems via Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.

Semia, security audit for AI agent skills.

Self-hosted runtime control plane for AI agents. Observe or HITL approve or Block rogue tool calls before it executes: secret leaks, prompt injection, supply chain etc in a local dashboard. Agent agnostic (Claude, codex, langchain etc.)

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

AI Agent Skill Security Scanner — Detects malicious code, prompt injection, data exfiltration, and supply chain threats in SKILL.md files. Built by Evolution Unleashed.

Composable security scanning harness for agent skills

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

Security scanner for AI agent skills and MCP tool bundles — prompt injection, IOC matching, malware detection, ML classifier

Static security scanner for SKILL.md files and AI agent skill packages. 25+ detection rules across 5 categories. Zero dependencies, CI/CD ready.

Security scanner for AI agent SKILL.md files. Static analysis, LLM behavioral prediction, and Docker Sandbox execution.

面向 Agent Skills 安全研究的检测绕过实验仓库，用于验证恶意 Skill 在静态扫描、语义分析与运行时执行之间可能存在的检测盲区。

A safer way to run autonomous CLI agents on your own computer. Open-source, MIT, community-driven.

Free agent security tools for OpenClaw -- ShellGuard Scanner + Token Audit

Scan agent skills and plugins for malware and prompt injection before an agent loads them. Offline GitHub Action and CLI. The published projection of OpenTrApp's Skill Firewall.

CLI to spot risky AI skill packages before you use them. Scans for injection, exfiltration, supply chain, and other threats.

Enable AI agents to autonomously create, evaluate, and evolve skills across any marketplace without user intervention.

Safety scanner for Claude Code SKILL.md files, scripts, templates, and references.

Scan-gated safety for Agent Skills before they reach Codex, Claude Code, CI, or a registry.

Security & trust vetting for AgentSkills: mandatory open-source scanner gate (Cisco, NVIDIA SkillSpector, Snyk, sentry), OWASP Top 10 for Agentic Applications mapping, and tiered review reports.

Find hidden instructions and prompt-injection patterns inside agent skill files - before they find you. 19 EN/RU regex rules, contextual whitelist, trusted-host scoring, zero dependencies.