Open Issues Need Help
View All on GitHub Add an MCP / third-party tool-trust deep dive 28 days ago
enhancement help wanted good first issue
Catch the prompt-injection exploit before it ships. A Claude Agent Skill that reviews LLM apps for injection and tool-abuse, traces untrusted input to dangerous sinks, and recommends architectural fixes. OWASP LLM Top 10.
#agent-skills#ai-security#appsec#claude-code#claude-skill#genai-security#llm#llm-security#owasp#prompt-injection#red-teaming#security
Add an insecure-output UI render (XSS) example 28 days ago
enhancement good first issue
Catch the prompt-injection exploit before it ships. A Claude Agent Skill that reviews LLM apps for injection and tool-abuse, traces untrusted input to dangerous sinks, and recommends architectural fixes. OWASP LLM Top 10.
#agent-skills#ai-security#appsec#claude-code#claude-skill#genai-security#llm#llm-security#owasp#prompt-injection#red-teaming#security
Add a multi-agent trust worked example 28 days ago
enhancement good first issue
Catch the prompt-injection exploit before it ships. A Claude Agent Skill that reviews LLM apps for injection and tool-abuse, traces untrusted input to dangerous sinks, and recommends architectural fixes. OWASP LLM Top 10.
#agent-skills#ai-security#appsec#claude-code#claude-skill#genai-security#llm#llm-security#owasp#prompt-injection#red-teaming#security