agentleFS
0xelitesystem0xelitesystem/claude-skills-collection/skills/changelog-writer/SKILL.md

changelog-writer

Skill0 starsChanged 5 months agoMITView on GitHub

Use when the user wants to generate human-readable changelog entries. Triggers on "write a changelog", "summarize this release", "what changed since v1.2", "release notes for this PR batch", or when the user provides a list of git commits, PRs, or a diff between two versions and wants prose output.

21 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use test-writer and readme-generator.

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.
  • leonardomsoSkill

    test-writer

    Generate comprehensive Vitest tests for code examples in JavaScript concept documentation pages, following project conventions and referencing source lines

    Also in 5 projects that use this

  • glinckerSkill

    readme-generator

    Generates comprehensive README.md files for software projects by analyzing codebase structure

    Also in 6 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 19 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If only commit subjects are provided, ask whether bodies/PRs are available, they're often where the "why" lives.

    line 18
  2. Distinctive:

    Get the audience: Default to "users of the software" (i.e.

    line 20
  3. Distinctive:

    MUST NOT: write more than one entry per atomic change.

    line 64
  4. Distinctive:

    No semantic versioning info available: ask what version this is for.

    line 73
  5. Distinctive:

    Commits with no clear category: put under "Changed" or ask the user.

    line 74
  6. Distinctive:

    Long change list (>30 entries): ask if you should aggressively trim or include everything.

    line 76

“Common” is measured: kept by at least 6 in 10 of the 19 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 Writer

Turn raw git history into a changelog humans actually read.

Process

  1. Get the input. Accept any of:

    • List of commit messages (git log v1.0..HEAD --oneline)
    • List of PR titles + descriptions
    • A merged diff between two tags/branches
    • A free-form list the user dictates

    If only commit subjects are provided, ask whether bodies/PRs are available, they're often where the "why" lives.

  2. Get the audience. Default to "users of the software" (i.e. external changelog). Ask if it's:
    • External: end users, customers. Skip refactors and internal changes. Lead with user-visible impact.
    • Internal: engineering team. Include refactors, infra changes, dependency bumps.
    • Hybrid: both. Use sections.
  3. Group entries into these sections in order:

    • Breaking changes (anything that requires consumer action)
    • Added (new features, new endpoints, new capabilities)
    • Changed (behavior changes that aren't breaking)
    • Fixed (bugs)
    • Deprecated (still works, will go away)
    • Removed (gone)
    • Security (CVEs, fixes for security issues)
    • Internal (only if audience is internal/hybrid)

    Skip sections with no entries; don't write empty headings.

  4. Write each entry as user-facing impact, not implementation.
    • Bad: "Refactored the AuthService to use DI"
    • Good: "Login flow now reports network errors instead of hanging"
    • Bad: "Bumped axios to 1.6.2"
    • Good: (skip for external; for internal: "axios 1.5 → 1.6.2 (security patch)")
  5. Format:
    • Use Markdown
    • One bullet per entry
    • Past tense ("Added", "Fixed") in the section heading; imperative or descriptive in the bullet
    • Reference PR/issue numbers in parens at the end of each bullet if available
    • Group related bullets together within a section
  6. Top of file: version + date heading.
    ## 2.4.0 - 2026-05-15
  7. Below the changelog, list anything you skipped and why. "Skipped 12 dependency bumps and 3 internal refactors." Lets the user catch you if you skipped something they wanted included.

What NOT to do

  • MUST NOT invent context. If a commit message says "fix bug", say "Fixed a bug", don't fabricate what the bug was.
  • MUST NOT include merge commits ("Merge branch X into Y") as changelog entries.
  • MUST NOT be vague. "Various improvements" is useless. Either name the improvement or omit.
  • MUST NOT include version-bump commits, lint-config commits, formatting-only commits, or CI-only commits in an external changelog.
  • MUST NOT write more than one entry per atomic change. If a feature shipped via 8 commits, that's one bullet.

Tone

  • External: clear, friendly, no jargon. A new user reading this should understand the impact.
  • Internal: dense, technical, assumes context. Engineers reading their own changelog don't need explanations.

Edge cases

  • No semantic versioning info available: ask what version this is for. Don't guess.
  • Commits with no clear category: put under "Changed" or ask the user.
  • Reverts: list as "Reverted X (introduced in Y)" with the original version.
  • Long change list (>30 entries): ask if you should aggressively trim or include everything. For external, almost always trim.

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.