More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    GitHub Actions Safety Check

    1

    Find the GitHub Actions mistakes that let a pull request or issue run code with your secrets: pull_request_target misuse, script injection, hard-coded tokens, unpinned actions and broad permissions.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    GitHub Actions Safety Check

    GitHub Actions Safety Check

    Example session with this skill installed

    Check this project with github-actions-safety-check.

    • Read your context and instructions
    • Compiled the github actions safety

    $ python3 scripts/check_actions.py examples/risky
    BLOCK deploy.yml:14 G03 HARDCODED_TOKEN
    curl -H "Authorization: Bearer xoxb..." https://slack.example.com/api/notify
    Why: A real access token (GitHub, GitLab, Slack or AWS) is written in the workflow file, so anyone who can read the repository can use it.
    Safer: Revoke the token now, then store a new one as a repository or environment secret and use ${{ secrets.NAME }}.

    BLOCK pr.yml:12 G01 PR_TARGET_RUNS_PR_CODE
    BLOCK pr.yml:15 G02 SCRIPT_INJECTION
    REVIEW deploy.yml:5 G13 NO_PERMISSIONS_SET
    REVIEW deploy.yml:10 G17 PIPE_TO_SHELL
    REVIEW deploy.yml:11 G14 SECRET_IN_RUN_SCRIPT
    REVIEW pr.yml:5 G12 WRITE_ALL_PERMISSIONS
    REVIEW pr.yml:8 G16 SELF_HOSTED_ON_PULL_REQUESTS
    REVIEW pr.yml:16 G11 UNPINNED_THIRD_PARTY_ACTION

    (Each finding also has Why and Safer lines, as in the first one.)
    Verdict: BLOCK. Checked 2 files, 2 with workflow content.
    This is a static check of file text. NOT FLAGGED is not a guarantee.

    BLOCK deploy.yml

    Connects securely to your tools. The creator never sees your data.

    About this skill

    An issue title pasted into a run script with ${{ }} can run any command in your workflow, with your secrets. This skill finds that, pull_request_target jobs that run a fork's code, and seven other workflow risks, and tells you how to fix each one.

    Your agent runs a scanner on the workflow files in .github/workflows (or one file you name). It returns BLOCK, REVIEW, or NOT FLAGGED, with the file, line, reason, and a safer approach for every finding.

    What it checks

    BLOCK

    • G01: a pull_request_target workflow that checks out the pull request's code
    • G02: issue, PR, comment or branch text pasted into a run script with ${{ }}
    • G03: a GitHub, GitLab, Slack or AWS token written in the workflow

    REVIEW

    • G11: a third-party action not pinned to a full commit SHA
    • G12: permissions: write-all
    • G13: no permissions block
    • G14: a secret pasted straight into a run script
    • G16: a self-hosted runner on pull request triggers
    • G17: a download piped into a shell

    What's included

    • The scanner (Python 3.9+, standard library only)
    • A rules guide with the reason and safer approach for each rule
    • Risky and safe example workflows, and the real report the scanner produces on the risky ones

    Limits It reads workflow YAML line by line. It does not resolve reusable workflows, composite actions or YAML anchors, and it cannot see repository settings. NOT FLAGGED is not a guarantee.

    Static, line-based check of workflow files. Reusable workflows, composite actions, anchors and repository or organisation settings are not resolved. NOT FLAGGED is not a guarantee.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 5 days ago

    Requires

    python3 (3.9 or newer)

    Permissions

    Terminal / Shell
    Read Files

    Runs one local Python script in the terminal. It reads your project's files, writes nothing, and makes no network calls. The example workflows include a clearly fake, inert token (xoxb-EXAMPLE-ONLY-0000) for the scanner to find.

    Listed8 days ago
    Updated5 days ago

    What's inside

    Frequently Asked Questions

    No. The script reads local files only and makes no network calls.

    No. It only reports. Your agent changes files only if you ask it to.

    Any agent that supports SKILL.md skills and can run a terminal command, such as Claude Code. It needs Python 3.9 or newer and nothing else.

    No. It reads the workflow YAML files. It does not run workflows, call the GitHub API or need a token.

    BLOCK: fix it before you ship. REVIEW: check it before release. NOT FLAGGED: no rule matched. It is a static check, so NOT FLAGGED is not a guarantee.

    The scanner, a rules guide with the reason and safer approach for every rule, a risky and a safe example, and the real report the scanner prints on the risky example.

    New in DevOps & Cloud