Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    release notes writer RN

    1

    Automate changelog generation and release announcements by analyzing git history and diffs.

    Free

    Security scanned
    release notes writer RN

    release notes writer RN

    Example session with this skill installed

    Generate release notes for the changes between v1.2.0 and the current master branch. Include a version suggestion.

    • Read your context and instructions
    • Compiled the release notes writer

    I have analyzed 14 commits and 3 merge requests between v1.2.0 and HEAD. I identified 4 user-facing changes and 10 internal updates (CI tweaks and dependency refactors) which were omitted from the public notes.

    Suggested Version: 1.3.0 (Minor) - New features added without breaking changes.

    Release Artifacts

    1. CHANGELOG.md Entry

    
    ## [1.3.0] - 2024-05-20
    ### Added
    - Search functionality for the project dashboard to find resources by tag.
    - Export to CSV option for the monthly usage reports.
    
    ### Fixed
    - Resolved a race condition during the OAuth login flow that caused occasional 500 errors.
    - Corrected button alignment on the mobile settings view.
    

    2. Announcement Blurb

    We are excited to release v1.3.0! This update introduces a highly requested search feature for the dashboard, making it easier to manage large projects. We have also added CSV exports for usage reports and fixed several UI bugs on mobile devices. No manual configuration changes are required for this update.

    3. Internal Summary

    This release covers 14 commits across 8 files. The primary risk area is the OAuth fix in auth-provider.ts, though unit tests pass. Dependency bumps for lodash and typescript were handled as internal noise. Rollback involves reverting to the v1.2.0 tag.

    Next steps

    • Review the classified changes to ensure no internal refactors were mislabeled as features.
    • Append the new section to your CHANGELOG.md file.
    • Tag the repository with git tag v1.3.0.
    1. CHANGELOG.md

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

    About this skill

    Translating raw git history into customer-ready documentation is a manual chore that often results in cryptic technical jargon. This skill automates the extraction, classification, and translation of commit logs into professional release notes and changelog entries. It filters out internal noise like CI tweaks and typo fixes to focus on what actually impacts the user.

    What it does

    • Evidence gathering analyzes git tags, commit logs, merge requests, and diff stats to build a complete picture of changes.
    • Noise reduction automatically identifies and excludes internal chores, reverts, and non-functional updates.
    • Semantic versioning suggests the next version bump (Major/Minor/Patch) based on the presence of breaking changes or new features.
    • Multi-format output generates structured CHANGELOG.md entries, concise announcement blurbs, and internal team summaries.
    • Change classification buckets updates into standard categories like Added, Fixed, Breaking, and Security.

    How it works

    1. Identify the range by finding the delta between the latest git tag and HEAD or a specific PR branch.
    2. Scan history using git log and diff commands to extract commit messages and file changes.
    3. Draft and classify by mapping technical commits to user-facing benefits while flagging breaking changes for upgrade notes.
    4. Review and refine by presenting the suggested version and a list of ignored "noise" commits for user approval.

    Frameworks & tools

    The skill requires a git-initialized repository and functions best in environments using Conventional Commits, though it can infer intent from standard messages.

    Why this beats prompting it yourself

    Writing release notes via generic prompts often results in hallucinated features or the inclusion of "internal-only" technical debt. This skill uses specific git commands to verify changes and follows the "Keep a Changelog" standard, ensuring your documentation remains accurate and professional.

    Use cases

    • Generating a pull request summary before merging a large feature branch.
    • Preparing a CHANGELOG.md update during the CI/CD tagging process.
    • Drafting a Discord or Slack announcement for a new production deployment.
    • Auditing a release for breaking changes that require specific migration guides.

    Known limitations

    Requires a git history to function. If no tags exist, it defaults to the first commit. It cannot verify if a "fix" actually works, only that the commit exists.

    Requires a git repository. Effectiveness depends on commit message quality. Cannot verify code functionality, only documents reported changes.

    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

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 23 days ago

    • Free with an Agensi account

    Requires

    Git

    Permissions

    Terminal / Shell
    Listed23 days ago

    What's inside

    Frequently Asked Questions

    This skill analyzes raw git logs and diffs to filter out 'internal noise' like typo fixes or CI updates, ensuring your public changelogs focus only on user-facing features and fixes.

    The skill functions best with Conventional Commits, but it is designed to infer intent and classify changes into categories like 'Added' or 'Fixed' even from standard commit messages.

    Yes, the skill automatically analyzes code changes for breaking changes and suggests whether the next release should be a Major, Minor, or Patch update according to Semantic Versioning.

    The skill requires a git-initialized repository and access to the command line to run git log and diff commands; it is compatible with any agent environment that has git installed.

    You receive the fully configured skill ready to integrate into your workflow, including templates for CHANGELOG.md files, Slack/Discord announcements, and internal summaries.

    New in Technical Documentation