LLM Agent Tooling Audit

Use this skill to audit LLM agents, tool calling, MCP integrations, prompt injection, data exfiltration, and tool permissions. Do not use it for ordinary backend authz unless agent/tooling is involved.

aibot88 Updated 3 repo stars

File contents

aibot88/sec_skill_store/tree/main/skills/claudskills/llm-agent-tooling-audit commit 10581de8ed

Frequently asked questions

npx skillmds@latest add aibot88/llm-agent-tooling-audit