agentleFS
degdeg/beads-utils/skills/changelog/SKILL.md

changelog

Skill2 starsChanged 6 days agoMITView on GitHub

Generate a changelog entry for recent work and write it into CHANGELOG.md, matching the file's existing style, one terse user-facing line per change. Use for /changelog, or when the user asks to write up what changed since the last release or tag. Leaves the entry unstaged for review and never commits.

166 other projects use a version of this.

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

9 rules. 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 155 edited versions we read keep these.

  • 7

    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:

    If the user specified a range (e.g., a tag, commit, or date), use that.

    line 12
  2. Distinctive:

    If no tags exist, use the first commit on the current branch diverging from main.

    line 14
  3. Very specific:

    Run git log <start>..HEAD --oneline to list all commits in range.

    line 19
  4. Distinctive:

    Read the existing CHANGELOG.md (if one exists) to match its format and style.

    line 21
  5. Distinctive:

    If CHANGELOG.md already exists, match its heading format and style exactly — including its category names, if it has its own.

    line 30
  6. Very specific:

    If no CHANGELOG.md exists, use format: ## vX.Y.Z (DDMonYY) with bullet points.

    line 31
  7. Distinctive:

    Write the new entry directly into CHANGELOG.md, inserting it above the most recent version block (or at the top if no CHANGELOG.md exists …

    line 35
  8. Distinctive:

    Do NOT git add or commit — leave the change unstaged until the user approves.

    line 37
  9. Distinctive:

    Every small UI tweak

    line 80

“Common” is measured: kept by at least 6 in 10 of the 155 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

Generate changelog entry

Generate a changelog entry for recent work and write it into CHANGELOG.md. Leave it unstaged for the user to review; never commit.

1. Determine the change range

  • If the user specified a range (e.g., a tag, commit, or date), use that.
  • Otherwise, auto-detect: find the most recent git tag with git describe --tags --abbrev=0.
  • If no tags exist, use the first commit on the current branch diverging from main.
  • Show the user what range you're using before proceeding.

2. Gather changes

  • Run git log <start>..HEAD --oneline to list all commits in range.
  • Also run git diff --stat <start>..HEAD to understand scope.
  • Read the existing CHANGELOG.md (if one exists) to match its format and style.

3. Draft the entry

Categorize commits and draft the entry following the conventions below. Apply editorial judgment:

  • Collapse related commits into a single line (e.g., "add feature X" + "fix feature X typo" = one entry).
  • Omit commits that are not user-visible or architecturally significant (see "What NOT to Include").
  • Order entries by significance within each category.
  • If CHANGELOG.md already exists, match its heading format and style exactly — including its category names, if it has its own.
  • If no CHANGELOG.md exists, use format: ## vX.Y.Z (DDMonYY) with bullet points.

4. Write the entry

  • Write the new entry directly into CHANGELOG.md, inserting it above the most recent version block (or at the top if no CHANGELOG.md exists yet).
  • Tell the user it has been written and where, so they can review and amend it in the file.
  • Do NOT git add or commit — leave the change unstaged until the user approves.

Changelog Conventions

Categories

The test is whether a user could notice the change. A user is anyone using the product from outside the code: every page of the app (admin pages included), reports, emails and exports, and the API. Logs, tests, scripts and the codebase itself are internal. Apply these in order:

  1. [bug-fix] — the old behaviour was wrong: incorrect or bad output, a failure, a crash. This holds whether or not users could see it, and takes precedence over the other two.
  2. [feature] — otherwise, a user can notice the difference: a new capability, an improvement to an existing one, a removal (prefix "Deprecation:" when it is one), a breaking change, or a visible change in output, wording, layout or speed.
  3. [cleanup] — otherwise, the change is internal: refactoring, dependency updates, tests, docs, logging, or a speed-up nobody would notice.

A visible change too small to be worth a line (a relabel, a restyle, a shortened title) is not filed as [cleanup]: it is dropped. A category says where a line goes, not whether it is included: the rules in "What NOT to Include" still apply.

Group the bullets by category in that output order — all [feature], then [bug-fix], then [cleanup] — never interleaved.

Principles

  1. Significance only: Major changes, not every small modification
  2. Order by impact: Most important first within each category
  3. User-facing perspective: Emphasize what affects users or architecture, in the names users see rather than internal or code terms ("Dashboard", not "dash-widget"), and say the outcome, not the mechanism
  4. One terse line per bullet: [category] verb object. No em-dash elaboration, no parenthetical detail dumps (rate limits, file names, migration numbers, version-by-version lists). "Add CSV export to the reports page" — not "Reports export endpoint — streamed, rate-limited, with optional column selection and a notification email". Multi-line entries are reserved for genuinely sprawling changes (e.g. a multi-finding security audit, a months-long overhaul) — ask before expanding.
  5. Prune hard: cut secondary clauses, and whole niche changes few readers care about

What NOT to Include

  • Minor bug fixes that don't affect functionality
  • Bug fixes for issues introduced in the same release
  • Internal refactoring that doesn't change behavior
  • Implementation details
  • Every small UI tweak
  • Routine maintenance (unless significant)

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.