メニュー
Use this skill to audit RAG and AI application security, including retrieval boundaries, prompt injection, citations, memory, and data exposure. Do not use it as a scanner or exploit runner.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use this skill to audit backend authorization, IDOR, ownership checks, and tenant isolation. Do not use it for smart contracts or legal contract review.
Use this skill to audit CI/CD workflows, dependencies, build scripts, releases, artifacts, and package publishing. Do not use it for runtime application authz review.
Use this skill only when the user explicitly selects security finding IDs to fix. Do not use it to fix all findings or perform broad refactors.
Use this skill to convert a security incident or public vulnerability pattern into reusable audit prompts, checklists, tests, and AGENTS.md rules. Do not use it to generate exploit instructions.
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.
Use this skill to review a PR or diff for security regressions. Do not use it for full-repository audits or legal contract review.
| name | rag-ai-app-security-audit |
| description | Use this skill to audit RAG and AI application security, including retrieval boundaries, prompt injection, citations, memory, and data exposure. Do not use it as a scanner or exploit runner. |
Audit RAG and AI application security in audit-only mode.
Use this skill when a review involves retrieval pipelines, document ingestion, embeddings, vector stores, citations, prompt construction, tool use, memory, tenant filters, or user-visible AI output.
Default to audit-only. Do not execute exploits, do not scan unrelated repositories, do not upload private source code or secrets, and Do not auto-merge. Ask for human approval before any patch that changes retrieval policy, tool permissions, memory, privacy, or authorization behavior.
以 audit-only 模式审计 RAG 和 AI application security。
当 review 涉及 retrieval pipelines、document ingestion、embeddings、vector stores、citations、prompt construction、tool use、memory、tenant filters 或用户可见 AI output 时，使用这个 skill。
默认 audit-only。不要执行 exploits，不要扫描无关仓库，不要上传私有源码或 secrets。Do not auto-merge。任何修改 retrieval policy、tool permissions、memory、privacy 或 authorization behavior 的 patch，都需要 human approval。
