Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    git commit pr writer

    1

    Standardize your Git history with diff-based Conventional Commits and structured PR descriptions.

    Free

    2 usesSecurity scanned

    git commit pr writer

    Example session with this skill installed

    Generate a commit message and PR description for my staged changes. I've updated the auth middleware to handle JWT expiration and added a unit test in auth_test.go.

    • Read your context and instructions
    • Compiled the git commit pr

    I have analyzed the staged changes in middleware/auth.go and middleware/auth_test.go. The diff shows a new logic block for token.Expired errors and a corresponding test case.

    Git Metadata & PR Description

    Commit Message:

    fix(auth): handle JWT expiration in middleware
    

    Pull Request Description:

    Summary

    This change updates the authentication middleware to explicitly catch and handle expired JSON Web Tokens. Previously, expired tokens could trigger a generic 500 error; they now return a specific unauthorized response.

    What changed

    • Added check for ErrExpiredToken in the Authorize middleware.
    • Updated the response header to include WWW-Authenticate: Bearer error="invalid_token".
    • Added a new test case in auth_test.go to verify expiration handling.

    Why

    Improving error clarity for frontend clients to trigger automatic re-authentication flows when a session expires.

    Impact & Risk

    • User Impact: Better UX; clients can now distinguish between "bad token" and "expired token".
    • Technical Impact: Minimal. Only affects the auth middleware layer.
    • Risk: Low. The change is isolated to error handling logic.

    Testing

    • Status: Verified (Based on new test coverage provided in diff).
    • Details: TestAuthMiddleware_ExpiredToken passes locally.

    Next steps

    1. Run git commit -m "fix(auth): handle JWT expiration in middleware"
    2. Verify if a corresponding ticket in Jira/GitHub needs to be linked.
    3. Push the branch and paste the PR description above into the repository UI.

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

    About this skill

    The problem

    Writing meaningful commit messages and PR descriptions is a repetitive tax on developer time. Manual summaries often miss subtle side effects, lack consistent formatting, or include "magic" claims about testing and intent that the code doesn't actually support.

    What it does

    • Analyzes staged diffs and branch comparisons to extract objective behavior changes.
    • Maps code modifications to strictly accurate Conventional Commit types like feat, fix, or refactor.
    • Identifies genuine breaking changes based on API removals and schema migrations rather than just diff size.
    • Generates structured PR descriptions including technical impact, risk assessment, and reviewer focus areas.
    • Flags unrelated changes in a single diff and suggests splitting them into distinct commits.

    Why this beats prompting it yourself

    Generic LLM prompts tend to hallucinate test results or invent issue numbers to fill templates. This skill enforces a "diff-as-evidence" policy, ensuring every claim in your PR is backed by actual code changes while strictly adhering to Conventional Commit standards without the fluff.

    Use cases

    • Generating standardized commit messages for teams enforcing Conventional Commits.
    • Drafting comprehensive PR descriptions for complex refactors or multi-file features.
    • Creating release notes by synthesizing a range of commits into a change summary.
    • Auditing staged changes to ensure no sensitive data or secrets are being committed.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.

    Will not invent issue links or test results; these must be provided by the user or environment. Primarily optimized for Conventional Commits.

    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

    2 uses

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed2 months ago

    What's inside

    Frequently Asked Questions

    The skill integrates with any Git-based environment and can be used through popular LLM interfaces that support file uploads or code snippets, such as ChatGPT, Claude, or terminal-based AI agents.

    Unlike generic prompts, this skill uses a "diff-as-evidence" approach that strictly maps code changes to Conventional Commit standards, preventing the AI from hallucinating details not present in the code.

    Yes, the skill is designed to identify multiple unrelated changes within a single diff and will suggest how to split them into distinct, logical commits for a cleaner history.

    You receive the specialized logic and structured templates required to transform raw diffs into standardized commits and PRs, along with instructions for integrating it into your existing workflow.

    Setting up is simple: just provide your staged changes or branch diff to the agent, and the skill will automatically begin the analysis and generation process based on its internal logic.

    The skill is designed to recognize API removals, schema migrations, and breaking syntax changes, ensuring that 'BREAKING CHANGE' footers are only added when the code actually warrants them.

    New in Code Quality & Review

    Free