agent-security-audit

Audit AI agent configurations for security risks — excessive permissions, prompt injection surfaces, data exfiltration paths, and missing guardrails. Use when reviewing CLAUDE.md files, MCP configs, agent orchestration code, or any AI agent setup.

Jump to install

Repository: OWASP/secure-agent-playbook
Last source activity: May 17, 2026 at 20:08
Detected SKILL.md language: English
Stars: 164
Forks: 24

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/OWASP/secure-agent-playbook --skill agent-security-audit

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

Agent Security Audit

Evaluate an AI agent's security posture by following the full procedure in plays/agent-security-audit.md.

Steps

Permission Inventory — Enumerate every tool, MCP server, file system path, network access, and credential the agent has. Flag capabilities beyond what its stated purpose requires.

Prompt Injection Surface Analysis — For each input path (user messages, tool outputs, MCP resources, RAG documents), assess whether crafted input could cause the agent to invoke unintended tools, override instructions, or exfiltrate data.

Excessive Agency Assessment (OWASP LLM06) — Check whether destructive/irreversible actions require confirmation, whether access exceeds need, whether the agent can escalate its own privileges, and whether individually-safe tool calls can chain into harmful outcomes.

Data Exfiltration Path Analysis — Map how sensitive data could leave the agent boundary: secrets passed to external tools, file contents in web requests, cross-MCP-server data forwarding, sensitive data in logs.

Tool-Call Injection Assessment — For each tool: can user-controlled input reach tool parameters unsanitized? Check for command injection, path traversal, SSRF, and SQL injection through agent-constructed calls.

Guardrail Evaluation — Check for system prompt safety instructions, tool call confirmations, output filtering, rate limiting, audit logging, and sandboxing.

Output

Use the finding format from templates/finding.md. Produce a Permission Summary table, Risk Findings, Injection Surface Map, and prioritized Recommendations.

OWASP References

LLM01: Prompt Injection

LLM02: Insecure Output Handling

LLM06: Excessive Agency

LLM07: System Prompt Leakage

LLM08: Vector and Embedding Weaknesses