Agent skill · Trail of Bits
Agentic Actions Auditor skill for Claude Code, Codex and Cursor
Officialname: agentic-actions-auditor★ 7.8k (repo)pushed 2026-10-24CC-BY-SA-4.0SKILL.md on GitHubchecked 2026-11-07
What it does, in the publisher's words
Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations including Claude Code Action, Gemini CLI, OpenAI Codex, and GitHub AI Inference. Detects attack vectors where attacker-controlled input reaches AI agents running in CI/CD pipelines, including env var intermediary patterns, direct expression injection, dangerous sandbox configurations, and wildcard user allowlists.
The description field of the SKILL.md. The agent reads only this at startup and decides from it when to load the full skill.
Install Agentic Actions Auditor
The skills CLI copies the skill into the folder of each agent you pick (Claude Code, Codex, Cursor, Copilot, Gemini CLI, OpenCode, Windsurf...). Add -g for all projects. Needs Node.js 22.20+. docs
npx skills add trailofbits/skills --skill agentic-actions-auditor
This repo is a plugin marketplace. Inside Claude Code, add it and install the plugin that holds the skill (audit-context-building, building-secure-contracts, burpsuite-project-parser, chrome-mcp-troubleshooting...), or copy only this skill into ~/.claude/skills. docs
/plugin marketplace add trailofbits/skills /plugin install <plugin>@trailofbits # or only this skill: npx skills add trailofbits/skills --skill agentic-actions-auditor -a claude-code
Download the skill folder, zip it, then Customize > Skills > + > Create skill > Upload a skill. Code execution must be on (Settings > Capabilities). docs
git clone --depth 1 https://github.com/trailofbits/skills.git /tmp/agentic-actions-auditor-src cp -r /tmp/agentic-actions-auditor-src/plugins/agentic-actions-auditor/skills/agentic-actions-auditor ./agentic-actions-auditor zip -r agentic-actions-auditor.zip agentic-actions-auditor
Codex reads skills from .agents/skills in the repo or ~/.agents/skills. Type $ in Codex to use one. docs
npx skills add trailofbits/skills --skill agentic-actions-auditor -a codex
gh skill (GitHub CLI 2.90+, preview) installs it for Copilot in VS Code, the CLI and the cloud agent. Copilot also reads .github/skills and ~/.copilot/skills. docs
gh skill install trailofbits/skills agentic-actions-auditor
Cursor reads .cursor/skills, .agents/skills and ~/.cursor/skills (and .claude/skills). It does not install loose skills from a GitHub URL, so copy it in. docs
npx skills add trailofbits/skills --skill agentic-actions-auditor -a cursor
Installs from the GitHub repo; --path picks this skill out of the repo. Default scope is user (all projects). docs
gemini skills install https://github.com/trailofbits/skills.git --path plugins/agentic-actions-auditor/skills/agentic-actions-auditor
OpenCode reads .opencode/skills, ~/.config/opencode/skills and the Claude and .agents folders. docs
npx skills add trailofbits/skills --skill agentic-actions-auditor -a opencode
Windsurf (now Devin Desktop) reads .devin/skills, .windsurf/skills and ~/.codeium/windsurf/skills. docs
npx skills add trailofbits/skills --skill agentic-actions-auditor -a windsurf
Works with
Claude Code, Claude.ai and Claude Desktop, Codex, GitHub Copilot, Cursor, Gemini CLI, OpenCode, Windsurf and other agents that read the open Agent Skills format.
- Pre-approves these tools (allowed-tools, experimental in the spec): Read Grep Glob Bash.
What's in the SKILL.md
- When to Use
- When NOT to Use
- Rationalizations to Reject
- Audit Methodology
- Step 0: Determine Analysis Mode
- Step 1: Discover Workflow Files
- Step 2: Identify AI Action Steps
- Step 3: Capture Security Context
- Step 4: Analyze for Attack Vectors
- Step 5: Report Findings
- Detailed References
Bundled files
- agents/openai.yaml
- assets/trail-of-bits-mark.svg
- references/action-profiles.md
- references/cross-file-resolution.md
- references/foundations.md
- references/vector-a-env-var-intermediary.md
- references/vector-b-direct-expression-injection.md
- references/vector-c-cli-data-fetch.md
- references/vector-d-pr-target-checkout.md
- references/vector-e-error-log-injection.md
- references/vector-f-subshell-expansion.md
- references/vector-g-eval-of-ai-output.md
- references/vector-h-dangerous-sandbox-configs.md
- references/vector-i-wildcard-allowlists.md
Loaded only when the instructions point to them.
Facts
- Skill name
- agentic-actions-auditor
- Publisher
- Trail of Bits (official)
- Repo
- trailofbits/skills
- License
- CC-BY-SA-4.0
- SKILL.md size
- 2,854 words, 320 lines
- Bundled files
- 14, 12 docs
- Repo stars
- 7.8k (as of 2026-10-24)
- SKILL.md checked
- 2026-11-07
More skills from trailofbits/skills
- CodeQL Scan: Builds a CodeQL database and runs data-flow and taint analysis to find vulnerabilities.
- Constant-Time Analysis: Finds timing side channels in cryptographic code, such as branches on secret values.
- Differential Security Review: Security review of a diff that weighs blast radius, call sites and test coverage of changed code.
- Modern Python Tooling: Sets up Python projects with uv, ruff and ty, or migrates from pip, Poetry and mypy.
- Property-Based Testing: Writes and debugs property-based tests with Hypothesis, fast-check, proptest and similar tools.
- Semgrep Rule Creator: Writes custom Semgrep rules to catch specific bug or vulnerability patterns.
- Sharp Edges: Flags error-prone APIs, risky defaults and footgun designs that invite security mistakes.
Related security & auth skills
- Security Best Practices Review · OpenAI
- Security Threat Model · OpenAI
- Sign In With Google (Web) · Google
- Security Review (Sentry) · Sentry
- Firebase Security Rules Auditor · Firebase (Google)
- Better Auth Best Practices · Better Auth
- Create Auth (Better Auth) · Better Auth
- Clerk Next.js Patterns · Clerk
FAQ
What does the agentic-actions-auditor skill do?+
Audits GitHub Actions workflows that run AI agents for prompt-injection and privilege risks. In its own SKILL.md, Trail of Bits describes it as: "Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations including Claude Code Action, Gemini CLI, OpenAI Codex, and GitHub AI Inference."
How do I install the agentic-actions-auditor skill?+
Run npx skills add trailofbits/skills --skill agentic-actions-auditor and pick your agents, or use the tab for your agent above. In Claude Code the skill ends up in ~/.claude/skills/agentic-actions-auditor/SKILL.md; in Codex and Cursor in .agents/skills/agentic-actions-auditor/.
Does Agentic Actions Auditor work in Codex, Cursor and Copilot, or only Claude?+
It's a standard SKILL.md, so any agent that reads the Agent Skills format can use it: Claude Code, the Claude apps, Codex, GitHub Copilot, Cursor, Gemini CLI, OpenCode, Windsurf and more.
Who maintains Agentic Actions Auditor?+
Trail of Bits, the company or project behind it, in the trailofbits/skills repo on GitHub, last pushed 2026-10-24. License: CC-BY-SA-4.0.
Read a skill's SKILL.md and scripts before you install it: skills aren't verified by any agent vendor. Claude skills marketplace · How to install skills · MCP servers