agentleFS
kasimmjkasimmj/claude-skills-mega/skills/changelog-gen/SKILL.md

changelog-gen

Skill1 starsChanged 6 months agoMITView on GitHub

Build a CHANGELOG.md from conventional commits. Groups by type (feat, fix, perf...) and infers semver bump.

  • Reads credentials

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All time: full history (rare.

    line 22
  2. Distinctive:

    If git describe fails (no tags), this is the first release — say so.

    line 24
  3. Very specific:

    If a CHANGELOG.md exists, prepend the new section above the most recent release.

    line 82
  4. Distinctive:

    If not, create one with a "Keep a Changelog" header.

    line 83
  5. Very specific:

    If package.json exists: npm version <bump> --no-git-tag-version

    line 90

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Changelog Generation Skill

You are generating a CHANGELOG entry from git history.

Step 1 — Pick the range

Ask (or infer):

  • Since the last tag: git log $(git describe --tags --abbrev=0)..HEAD
  • Since a specific tag: git log v1.2.0..HEAD
  • All time: full history (rare; only for first release)

If git describe fails (no tags), this is the first release — say so.

Step 2 — Parse conventional commits

Recognized prefixes:

  • feat: / feat(scope): → Features
  • fix: / fix(scope): → Bug Fixes
  • perf: → Performance
  • refactor: → Refactors
  • docs: → Documentation
  • test: → Tests
  • build: / ci: → Build & CI
  • chore: → Chore (often omitted from changelog)
  • revert: → Reverts

Breaking changes:

  • Footer BREAKING CHANGE: or ! after type (e.g., feat!:)
  • These get a 🚨 prefix and forces a major bump

Step 3 — Infer semver bump

Commits foundBump
Any BREAKING CHANGEMAJOR
Any feat: (no breaking)MINOR
Only fix:, perf:, refactor:, docs:, etc.PATCH
Only docs:, chore:, test:No release recommended

Step 4 — Write the CHANGELOG

Format:

## [1.4.0] — 2026-05-22

### 🚀 Features
- **auth**: support OAuth via GitHub (#142)
- **dashboard**: add export-to-CSV button (#138)

### 🐛 Bug Fixes
- **api**: handle 429 from upstream gracefully (#145)

### ⚡ Performance
- **db**: index `users.email` (#147) — login query 12× faster

### 🚨 Breaking Changes
- **config**: rename `API_KEY` → `OPENAI_API_KEY` (#150)
  Migration: rename your env var. See UPGRADE.md.

Each entry:

  • Includes the scope if present
  • Links the PR or commit hash
  • Strips the conventional prefix (no feat: in the bullet itself)
  • Has a one-line description — full PR title

Step 5 — Update CHANGELOG.md

If a CHANGELOG.md exists, prepend the new section above the most recent release. If not, create one with a "Keep a Changelog" header.

Step 6 — Suggest next steps

Print:

  • New version: <computed>
  • Suggested tag: git tag -a vX.Y.Z -m "Release vX.Y.Z"
  • If package.json exists: npm version <bump> --no-git-tag-version

When NOT to use

  • Commits don't follow conventional commits — recommend cz-commitlint setup first
  • The user wants release notes in a totally different format (e.g., for marketing)

Failure modes

  • ⚠️ Squash-merged PRs may lose individual commit messages. The PR title becomes the commit. That's usually fine.
  • ⚠️ If a commit message is just fix: stuff, flag it — recommend amending.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.