GitHub Actions GITHUB_TOKEN Permission Diff

    1

    Compare two workflow YAML files and review declared GITHUB_TOKEN permission changes before merge.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    GitHub Actions GITHUB_TOKEN Permission Diff

    Example session with this skill installed

    Compare the included invented before.yml and after.yml workflow files. Show the declared permission change at the publish job and the checks a maintainer should do before merging. Do not call GitHub or claim effective token privileges.

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

    Synthetic GitHub Actions permission change card

    Scope: only the supplied before.yml and after.yml workflow declarations.

    Status: REVIEW. The examples are invented and no GitHub API was called.

    LocationBefore declarationAfter declarationClassification
    job:publishcontents: readcontents: writeDECLARED_BROADENING

    The workflow-level contents: read declaration remains unchanged. The publish job has its own explicit permission map, so the card highlights that job-level change rather than assuming a live token grant. The output packet includes the before/after file hashes and an integrity receipt so reviewers can tie the card to the exact submitted files.

    Human review queue: Confirm these are the intended versions. Inspect the changed job scope and whether any other job overrides the workflow declaration. Check repository and organization defaults, event and fork behavior, reusable workflows and actual runtime token restrictions separately. Decide whether contents: write is justified for the publish job before merging.

    Limits: This is a static comparison of two supplied YAML files. DECLARED_BROADENING is a prompt for review, not proof that the job can write to the repository. The tool does not execute workflows, connect to GitHub, inspect secrets, or change settings.

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

    About this skill

    A small workflow edit can broaden declared GITHUB_TOKEN permissions at the workflow or job level. This local checker compares before-and-after YAML files and creates a deterministic change card, review checklist and integrity receipt. It does not connect to GitHub, execute a workflow or determine effective repository permissions.

    Give the included local CLI or optional stdio MCP tool before and after copies of one GitHub Actions workflow. It compares top-level and per-job `permissions`, flags changed and potentially broadened declarations, and writes a deterministic change card, human checklist and integrity receipt. The companion SKILL.md guides an assistant through reviewing that card. The ZIP includes readable Python source, a CLI, optional MCP adapter, invented YAML examples and sample output. Python 3.11–3.14 and pinned PyYAML 6.0.3 are required; optional MCP mode also uses mcp==2.2.0. At runtime it does not connect to GitHub, execute jobs, edit workflows or require a token. This is a review queue for declarations in the supplied files, not proof of effective GITHUB_TOKEN rights, security or repository defaults.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.

    Only declarations in the two supplied YAML files are compared. Repository and organization defaults, event/fork behavior, reusable workflows and runtime token restrictions are not evaluated. A declared broadening is a review cue, not proof of effective privilege escalation or workflow security.

    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 buying a skill.

    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 1 month ago

    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    **/*

    Reads only the two local workflow YAML files selected by the buyer and writes a new user-selected result folder. The broad file glob reflects user-selectable paths, not background access. No GitHub account, API token, browser or network access is used at runtime; docs.github.com in the README is only a reference.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill analyzes and compares two versions of a GitHub Actions YAML file to detect changes in `GITHUB_TOKEN` permissions. It helps developers prevent 'permission creep' by providing a clear change card and checklist for any expanded access rights.

    The package includes the Python source code, a CLI tool, an optional MCP adapter for AI agent integration, sample YAML files, and a SKILL.md guide for assistant-led reviews.

    Yes, the skill includes a standard Model Context Protocol (MCP) adapter, making it compatible with any AI agent or IDE (like Claude Desktop) that supports the MCP standard.

    The tool is designed for privacy and local security; it only analyzes the local YAML files you provide and does not connect to the GitHub API or require a GitHub token to function.

    You will need Python 3.11–3.14 and the PyYAML library. If you plan to use the MCP integration, the `mcp` Python package is also required.

    This tool focuses strictly on reviewing the permission declarations within your workflow files; it does not verify organization-level defaults or check if a job successfully executes with those rights.

    New in Security