More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Git Manager.

    3

    Git manager stops your coding agent from committing to main, staging secrets, or losing uncommitted work. Handles branches, commits, and pull requests on its own the rest of the time.

    $10

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Git Manager.

    Git Manager.

    Example session with this skill installed

    I've finished the auth fix. Review the diff in auth.ts, stage only the relevant hunks, and commit it with a clear message. Do not push yet.

    • Read your context and instructions
    • Compiled the git manager

    I have inspected the repository state. HEAD is at 8a2f3c1 on branch feat/auth-fix.
    Staged 2 hunks in src/auth.ts covering the token refresh logic.
    Verified invariants: 12 unrelated files remain unstaged.
    Commit created: 5b1d9e2 fix(auth): resolve token refresh race condition

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

    About this skill

    The problem

    Coding agents commit straight to main, force-push over a protected branch, or run git add . and sweep a .env file in with everything else. None of that is hypothetical. It's what happens when an agent handles Git without something checking what it's about to do before it does it.

    What it does

    • Path-scoped staging with hunk review, so git add . accidents can't happen.
    • Blocks direct commits to main/master, so protected trunk stays protected.
    • Checks repository identity and worktree state before every mutation, not just once at the start.
    • Confirms remote state through the GitHub CLI (gh) instead of trusting a local exit code.
    • Gives you a disclosed recovery plan before history-changing operations like interactive rebase or conflict resolution, not after something's already gone sideways.

    Frameworks & tools

    Standard git and gh (GitHub CLI). Built for CLI-integrated environments like Claude Code or Cursor.

    Why this beats prompting it yourself

    Telling an agent to "be careful with git" doesn't cover the edge cases: detached HEAD, a rebase already in progress, a branch that moved since the plan was made. This skill re-checks HEAD and the index immediately before executing, so a plan from a few messages ago can't collide with changes you made in between.

    Use cases

    • Feature branch work that won't quietly touch main.
    • Interactive rebase and conflict resolution with a recovery step already planned out.
    • Pull requests that get verified against GitHub's actual state before they're reported done.
    • Staging exactly the hunks you meant to, without dragging in unrelated local changes.

    Known limitations

    Needs standard git and gh binaries installed. Won't run repository-provided scripts, hooks, or custom aliases. That boundary is intentional, not a gap.

    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 1 month ago

    Listed2 months ago
    Updated1 month ago

    What's inside

    Frequently Asked Questions