# GitHub Actions for AI agents (AgentGid)

Checked 2026-11-16. Facts from each Action's action.yml, latest release and repository. used_in = public repos set up for AI coding agents whose workflows use it (7,205 scanned).

## Mention an agent in an issue or pull request and get the change

- Claude Code Action v1.0 (`uses: anthropics/claude-code-action@v1`): Flexible GitHub automation platform with Claude. Auto-detects mode based on event type: PR reviews, @claude mentions, or custom automation. Secrets in inputs: anthropic_api_key, claude_code_oauth_token, github_token, ssh_signing_key. Used in 131 agent projects. https://github.com/anthropics/claude-code-action
- Codex Exec Action (`uses: openai/codex-action@<version>`): Run `codex exec` with a prompt. Secrets in inputs: openai-api-key. Used in 11 agent projects. https://github.com/openai/codex-action
- AI Inference (`uses: actions/ai-inference@v3`): Generate an AI response based on a provided prompt. Secrets in inputs: token. Used in 6 agent projects. The current action.yml supports only the copilot provider, which runs the GitHub Copilot CLI. The repository still describes it as calling models with GitHub Models, a service GitHub retired on July 30, 2026: check which version your workflow pins. https://github.com/actions/ai-inference
- Run Gemini CLI (`uses: google-github-actions/run-gemini-cli@v0.1.22`): Invoke the Gemini CLI from a GitHub Action. Secrets in inputs: gemini_api_key, google_api_key. Used in 5 agent projects. https://github.com/google-github-actions/run-gemini-cli
- Claude Code Base Action (`uses: anthropics/claude-code-base-action@<version>`): Run Claude Code in GitHub Actions workflows. Secrets in inputs: anthropic_api_key, claude_code_oauth_token. https://github.com/anthropics/claude-code-base-action
- Jules Invoke (`uses: google-labs-code/jules-action@v1.0.0`): Invoke Jules, an AI-powered coding agent, to perform tasks on your codebase. Secrets in inputs: jules_api_key. https://github.com/google-labs-code/jules-action
- Run an Oz cloud agent (`uses: warpdotdev/oz-agent-action@v1.0.37`): Run the Oz agent in a GitHub workflow. Secrets in inputs: warp_api_key. https://github.com/warpdotdev/oz-agent-action
- cagent Runner (`uses: docker/cagent-action@v1.5.5`): Run a cagent AI agent with a single line. Secrets in inputs: anthropic-api-key, openai-api-key, google-api-key, aws-bearer-token-bedrock, xai-api-key, nebius-api-key, mistral-api-key, github-token, org-membership-token. https://github.com/docker/cagent-action

## Get an AI review on every pull request

- The PR Agent (`uses: The-PR-Agent/pr-agent@v0.50.0`): Summarize, review and suggest improvements for pull requests. https://github.com/The-PR-Agent/pr-agent
- ChatGPT CodeReviewer (`uses: anc95/ChatGPT-CodeReview@v1.0.24`): A Code Review Action Powered By ChatGPT. https://github.com/anc95/ChatGPT-CodeReview
- Shippie 🚢 (`uses: mattzcarey/shippie@v0.21.2`): an extensible code review agent built on flue. Secrets in inputs: ANTHROPIC_API_KEY, OPENAI_API_KEY, OPENROUTER_API_KEY, CLOUDFLARE_API_KEY, GITHUB_TOKEN. https://github.com/mattzcarey/shippie
- OpenQodex AI Code Review: SAST, Secret Scanning and Lint for Pull Requests (`uses: openqodex/openqodex@<version>`): AI code review and scanners for pull requests: SAST, secrets, dependencies, lint. Review needs your Anthropic key. https://github.com/openqodex/openqodex
- AI Code Review Action (`uses: villesau/ai-codereviewer@v2.7.0`): Perform code reviews and comment on diffs using OpenAI API. Secrets in inputs: GITHUB_TOKEN, OPENAI_API_KEY. https://github.com/villesau/ai-codereviewer
- GenAI Code Review (`uses: cirolini/genai-code-review@v3.0.1`): Reviews pull request diffs with an LLM (Gemini, OpenAI, Anthropic or any OpenAI-compatible server) and posts only the few inline comments most worth a human's attention, saying what it held back and what it did not review. Secrets in inputs: api_key, openai_api_key, github_token. https://github.com/cirolini/genai-code-review
- Presubmit - AI Code Review & Fixes (`uses: presubmit/ai-reviewer@v0.2.5`): AI-powered code reviews that detect issues, provide summaries, and suggest auto-fixes on PRs. https://github.com/presubmit/ai-reviewer
- Juror AI (`uses: Juror-AI/juror@v1.4.8`): Multi-model PR review that shows you the bill — duplicate findings collapse into one, with optional agreement filtering. Secrets in inputs: github-token. https://github.com/Juror-AI/juror
- Sourcery Action (`uses: sourcery-ai/action@v1`): A GitHub Action for running Sourcery. Secrets in inputs: token. https://github.com/sourcery-ai/action

## Check AI-written code and agent setups for security problems

- HOL AI Plugin Scanner (`uses: hashgraph-online/ai-plugin-scanner-action@v1.2.817`): Scan AI plugin repositories for security, publishability, and best-practice quality signals. Secrets in inputs: submission_token. Used in 6 agent projects. https://github.com/hashgraph-online/ai-plugin-scanner-action
- OpenAI Codex Security (`uses: openai/codex-security@<version>`): Scan code with Codex Security, enforce findings policy, and produce JSON and SARIF reports. https://github.com/openai/codex-security
- Claude Code Security Reviewer (`uses: anthropics/claude-code-security-review@<version>`): AI-powered security review GitHub Action using Claude to analyze code changes for security vulnerabilities. Secrets in inputs: claude-api-key. https://github.com/anthropics/claude-code-security-review
- Pipelock Agent Security Scan (`uses: pipelab-org/pipelock@v3.8.0`): Scan your project for AI agent security risks. Pattern-based detection of credentials and misconfigurations with tailored config generation. See docs for coverage details. https://github.com/pipelab-org/pipelock
- Skylos - Python SAST, Dead Code Detection & PR Gate (`uses: duriantaco/skylos@v5.6.0`): SAST, dead code, secrets, PR gates, and digest-pinned container image vulnerability scans. Secrets in inputs: token. https://github.com/duriantaco/skylos
- aislop — AI Code Quality Gate (`uses: scanaislop/aislop@v0.22.1`): Run the aislop quality gate on AI-assisted code (scan + fail CI below threshold from .aislop/config.yml). https://github.com/scanaislop/aislop
- Vet: Verify Everything (`uses: imbue-ai/vet@v0.2.16`): LLM-based code review that finds issues tests and linters miss. https://github.com/imbue-ai/vet
- AI-BOM Scan (`uses: Trusera/ai-bom@v3.1.0`): Discover and inventory AI/LLM agents, models, and API integrations in your codebase. Generates SBOM-style reports in CycloneDX, SARIF, SPDX, and more. https://github.com/Trusera/ai-bom
- node9 Agent Security (`uses: node9-ai/node9-proxy@v2.62.1`): Scan a repo for agent-security risks (injectable CI workflows, unsafe agent configs, MCP) and comment on the PR. Secrets in inputs: github-token. https://github.com/node9-ai/node9-proxy
- LintLang Agent Config Linter (`uses: hermes-labs-ai/lintlang@v0.8.3`): Statically lint AI agent configs, tool definitions, and prompts with LintLang. https://github.com/hermes-labs-ai/lintlang
- AI Surface Check (`uses: apisec-inc/AI-Surface@v1.1.1`): Map AI attack surfaces in pull requests and gate new high-risk findings before merge. Secrets in inputs: github-token. https://github.com/apisec-inc/AI-Surface

## Test prompts, agents and AI features on every change

- Setup aimock (`uses: CopilotKit/aimock@v1.48.0`): Start an aimock server for AI application testing — LLM APIs, MCP, A2A, AG-UI, vector DBs. https://github.com/CopilotKit/aimock
- mcpsnoop (`uses: kerlenton/mcpsnoop@v0.28.0`): Fail CI on what an MCP server actually did on the wire, and file every finding as a code scanning alert. https://github.com/kerlenton/mcpsnoop
- coder_eval (`uses: UiPath/coder_eval@v0.12.18`): Install a pinned coder-eval and run evaluation tasks as a CI gate, with JUnit XML output. https://github.com/UiPath/coder_eval
- Azure AI Agent Evaluation (`uses: microsoft/ai-agent-evals@<version>`): Evaluate the performance, quality and safety of AI agents using the Azure AI Evaluation SDK. https://github.com/microsoft/ai-agent-evals
- Test LLM outputs (`uses: promptfoo/promptfoo-action@<version>`): Run promptfoo evaluations on your prompts. Works with pull requests, pushes, and manual triggers (workflow_dispatch). Secrets in inputs: github-token, openai-api-key, azure-api-key, anthropic-api-key, huggingface-api-key, aws-access-key-id, aws-secret-access-key, replicate-api-key, palm-api-key, vertex-api-key, cohere-api-key, mistral-api-key, groq-api-key. https://github.com/promptfoo/promptfoo-action
- ollama-action (`uses: ai-action/ollama-action@v2.0.1`): Run Ollama large language models (LLMs) with GitHub Actions. https://github.com/ai-action/ollama-action
- Braintrust eval (`uses: braintrustdata/eval-action@v2.1.0`): Automatically run evals on your AI project using Braintrust. Secrets in inputs: api_key, github_token. https://github.com/braintrustdata/eval-action
- Tessl Skill Review (`uses: tesslio/skill-review@<version>`): Automatically review SKILL.md files changed in a pull request and post results as a PR comment. Secrets in inputs: tessl-token. https://github.com/tesslio/skill-review

## Translate the app and fix accessibility automatically

- Lingo.Dev AI Localization (`uses: lingodotdev/lingo.dev@<version>`): Automated AI localization for dev teams. Secrets in inputs: api-key. https://github.com/lingodotdev/lingo.dev
- Languine (`uses: languine-ai/languine@v1`): Self-hosted automated localization powered by your own Languine deployment. Secrets in inputs: api-key. https://github.com/languine-ai/languine
- accessibility-scanner (`uses: github/accessibility-scanner@v3.5.2`): Finds potential accessibility gaps, files GitHub issues to track them, and attempts to fix them with Copilot. Secrets in inputs: token, password. https://github.com/github/accessibility-scanner

Pin third-party Actions to a full-length commit SHA and give workflows only the permissions they need (https://docs.github.com/en/actions/reference/security/secure-use).
