HeadyZhang

HeadyZhang/agent-audit

Static security scanner for LLM agents — prompt injection, MCP config auditing, taint analysis. 51 rules mapped to OWASP Agentic Top 10 (2026). Works with LangChain, CrewAI, AutoGen.

192 stars 22 forks 192 watchers Python MIT License
ai-agent ai-security ai-security-tool cli crewai langchain langchain-security- llm-safety llm-security mcp owasp prompt-injection python security static-analysis taint-analysis vulnerability-detection vulnerability-scanner
View on GitHub Website
1 Open Issue Need Help Last updated: Jul 4, 2026

Open Issues Need Help

View All on GitHub
False positive: AGENT-041 flags agent `.execute()` calls as SQL injection about 2 hours ago
bug help wanted
HeadyZhang/agent-audit
192

Static security scanner for LLM agents — prompt injection, MCP config auditing, taint analysis. 51 rules mapped to OWASP Agentic Top 10 (2026). Works with LangChain, CrewAI, AutoGen.

Python
#ai-agent#ai-security#ai-security-tool#cli#crewai#langchain#langchain-security-#llm-safety#llm-security#mcp#owasp#prompt-injection#python#security#static-analysis#taint-analysis#vulnerability-detection#vulnerability-scanner