C

Changelog Generator Toolkit

Comprehensive skill designed for automatically, creates, user, facing. Includes structured workflows, validation checks, and reusable patterns for development.

SkillClipticsdevelopmentv1.0.0MIT
0 views0 copies

Changelog Generator

A Claude Code skill that transforms raw git commit history into polished, user-friendly changelogs organized by semantic versioning, categorized by change type, and formatted for your audience.

When to Use This Skill

Choose this skill when:

  • Preparing release notes for a new software version
  • Creating weekly or monthly product update summaries
  • Generating changelogs for open-source project releases
  • Automating CHANGELOG.md maintenance in CI/CD pipelines
  • Communicating technical changes to non-technical stakeholders

Consider alternatives when:

  • You need a full release management system (use semantic-release or release-please)
  • You need real-time commit notifications (use a CI webhook instead)
  • Your project uses conventional commits with an existing changelog tool already configured

Quick Start

# Add to your Claude Code project claude mcp add changelog-generator # Generate changelog for the latest release claude "generate changelog from v1.2.0 to HEAD" # Generate user-friendly release notes claude "create release notes for version 2.0 targeting end users"
# .claude/skills/changelog-generator.yml name: changelog-generator description: Transform git commits into polished changelogs triggers: - "generate changelog" - "release notes" - "what changed" config: format: keep-a-changelog audience: developers group_by: type

Core Concepts

Change Categories

CategoryCommit PrefixesDescription
Addedfeat:, add:New features and capabilities
Changedchange:, update:, refactor:Modifications to existing features
Deprecateddeprecate:Features marked for future removal
Removedremove:, delete:Removed features or capabilities
Fixedfix:, bugfix:Bug fixes and corrections
Securitysecurity:, vuln:Security patches and vulnerability fixes
Performanceperf:Performance improvements
Documentationdocs:Documentation updates

Output Example

# Changelog ## [2.1.0] - 2026-03-13 ### Added - User profile avatars with automatic image optimization (#234) - Dark mode support across all dashboard pages (#251) - Export data to CSV and PDF formats (#267) ### Fixed - Search results no longer return duplicate entries (#289) - Password reset emails now work with custom domains (#293) ### Changed - Upgraded authentication flow to OAuth 2.1 standard (#278) - Improved table loading performance by 40% with virtual scrolling (#281) ### Security - Patched XSS vulnerability in comment rendering (#301)

Configuration

ParameterTypeDefaultDescription
formatstring"keep-a-changelog"Output format: keep-a-changelog, conventional, custom
audiencestring"developers"Target audience: developers, users, stakeholders
group_bystring"type"Group entries by: type, scope, component
include_commitsbooleanfalseInclude raw commit hashes in entries
include_authorsbooleanfalseInclude commit author names
include_prsbooleantrueLink to pull request numbers
exclude_typesarray["chore", "ci", "test"]Commit types to exclude from output
date_formatstring"YYYY-MM-DD"Date format for version headers
max_entriesnumber100Maximum entries per release

Best Practices

  1. Use conventional commit format consistently — the skill works best when commits follow patterns like feat: add search or fix: resolve login bug; inconsistent messages produce noisy changelogs.

  2. Write changelogs for your audience, not yourself — if the audience is end users, translate "refactored UserService dependency injection" into "improved account page loading speed."

  3. Group related changes under a single entry — five commits fixing the same feature should become one changelog entry, not five separate lines that obscure the bigger picture.

  4. Tag releases in git before generating — the skill uses git tags as version boundaries; without tags, it cannot determine which commits belong to which release.

  5. Review and edit generated changelogs — automated generation provides a strong starting point but human judgment is needed to highlight the most important changes and add context.

Common Issues

Non-conventional commits produce "Uncategorized" entries — If your team doesn't follow conventional commit format, configure custom prefix mappings or use the audience: users mode which uses AI to categorize commits by content rather than prefix.

Merge commits create duplicate entries — The skill may count both the merge commit and the individual commits. Set exclude_merges: true or use squash merging in your workflow to avoid duplicates.

Missing version tags cause single massive changelog — Without git tags marking releases, all commits appear under one version. Run git tag v1.0.0 <commit-hash> on historical release points before generating, or use --from and --to flags to specify commit ranges manually.

Community

Reviews

Write a review

No reviews yet. Be the first to review this template!

Similar Templates