agentleFS
definite-ddefinite-d/agent-skills/skills/changelog/SKILL.md

changelog

Skill2 starsChanged 4 months agoMITView on GitHub

Write well-formatted changelogs following Keep a Changelog and SemVer, from scratch or from existing records (git history, tags, release notes, an existing CHANGELOG.md). Buckets changes into Added/Changed/Deprecated/Removed/Fixed/Security, derives the next version with a semver bump, and preserves the file's existing conventions when updating. Use proactively when generating, regenerating, or maintaining a CHANGELOG.md, or drafting release notes.

160 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use playtest-report, soak-test and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • CoralGameSkill

    playtest-report

    Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.

    Also in 12 projects that use this

  • lilangMaxSkill

    soak-test

    Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.

    Also in 9 projects that use this

  • lilangMaxSkill

    team-live-ops

    Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.

    Also in 9 projects that use this

  • CoralGameSkill

    team-polish

    Orchestrate the polish team: coordinates performance-analyst, technical-artist, sound-designer, and qa-tester to optimize, polish, and harden a feature or area for release quality.

    Also in 14 projects that use this

  • +5 more from lilangMax
  • +3 more from CoralGame

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 149 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep a Changelog categories, ordered as shown:

    line 51
  2. Distinctive:

    Write user-facing summaries, not raw commit messages.

    line 64
  3. Very specific:

    Read references/version-bumping.md for the full decision tree.

    line 72
  4. Very specific:

    No prior release → start at 0.1.0 (or 1.0.0 if the project is stable by its own convention)

    line 79
  5. Very specific:

    Never rewrite released sections.

    line 118

“Common” is measured: kept by at least 6 in 10 of the 149 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 Skill

Produce human-readable, release-ready changelogs that follow Keep a Changelog structure and Semantic Versioning bump rules, whether starting from an empty repo or maintaining an existing file.

Mode Selection

ModeTriggerAction
From scratchNo CHANGELOG.md existsBuild full file from git history / records
Update existingCHANGELOG.md existsMerge new changes, preserving its conventions
SubsetUser names a version rangeExtract only that window

Confirm the mode and target (whole repo, a package, a file) before writing.

Gather Records

Prioritize sources in this order:

  1. Existing changelog — parse headings, section names, date format, link style. Preserve them.
  2. Git history — scope per mode:
    git tag --sort=-version:refname              # release anchors
    git log <last_tag>..HEAD --oneline --decorate # unreleased changes
    git log --oneline --decorate                 # full history (from scratch)
    git log -1 --format='%ad %h' --date=short <tag>  # release date for a tag
  3. Release records — release notes, PR descriptions, issues closed in a milestone. Treat as ground truth when present; use commits to fill gaps.
  4. User-provided lists — commit messages pasted in chat. Parse them as-is.

Skip merge commits, dependabot noise, and generated files unless they affect users.

Bucket Into Standard Sections

Keep a Changelog categories, ordered as shown:

SectionConventional type mappingExample entry
Addedfeat, perf (new)- **Added:** rate_limit middleware for public endpoints.
Changedrefactor, perf, style- **Changed:** JSON serialization to use orjson.
Fixedfix- **Fixed:** panic on empty payload in the webhook handler.
Removedremove, revert, deprecate follow-up- **Removed:** legacy v1 auth endpoints.
Deprecateddeprecate- **Deprecated:** --envflag; use--profile instead.
Securityfix (CVE), security- **Security:** patched SSRF in proxy URL validation.

Rules:

  • Write user-facing summaries, not raw commit messages. Imperative or past tense is fine, but stay consistent within one file.
  • One bullet per logical change; fold related commits (same feature, multiple parts) into a single entry.
  • Reference issue/PR numbers ([#42](https://.../pull/42)) only if the project already does so.
  • Omit empty sections; keep the headings exactly ### Added, ### Changed, etc.
  • Mark breaking changes inside the relevant section with a **BREAKING:** prefix and call them out in the release summary.

Determine the Next Version

Read references/version-bumping.md for the full decision tree.

Quick rules (from the last released tag):

  • Breaking change → bump major (or minor when 0.x)
  • New user-facing feature → bump minor (or patch when 0.x)
  • Fixes/refactors/docs → bump patch
  • No prior release → start at 0.1.0 (or 1.0.0 if the project is stable by its own convention)

Write the File

From scratch — Keep a Changelog skeleton

# Changelog

All notable changes to this project will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## [Unreleased]

### Added

- ...

## [1.0.0] - 2026-07-31

### Fixed

- ...

[Unreleased]: https://github.com/OWNER/REPO/compare/v1.0.0...HEAD
[1.0.0]: https://github.com/OWNER/REPO/compare/v0.1.0...v1.0.0

Format rules:

  • ## [Unreleased] always first; released versions newest-first.
  • Version headings: ## [X.Y.Z] - YYYY-MM-DD (ISO date, - separators).
  • Resolve OWNER/REPO from git remote get-url origin; omit the diff-link footer if no remote exists.
  • First release link points at the tag itself, not a comparison.

Updating an existing file

  • Reuse its exact heading depth, date format, link style, and intro line.
  • Never rewrite released sections. Only maintain [Unreleased] or append a new version heading.
  • Move the previous [Unreleased] content into the new release heading on release, then start a fresh [Unreleased].
  • Deduplicate: if an entry already appears under a released version, do not copy it forward.

Verify

  • Version numbers in headings match the footer links exactly (with/without v prefix consistent).
  • Dates are YYYY-MM-DD.
  • Sections present in canonical order, empty ones omitted.
  • Newest release on top, [Unreleased] above all releases.
  • No duplicate entries across sections or versions.
  • Semver bump matches content (breaking → major, feature → minor, fix → patch).
  • Repo-specific conventions preserved (scope, link format, trailing punctuation).

See Also

  • references/keep-a-changelog.md — Full format spec and edge cases
  • references/version-bumping.md — SemVer decision tree with 0.x and prerelease rules

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.