agentleFS
khadir-syedkhadir-syed/k_ai-agent-skills/skills/software/changelog-entry-drafter/SKILL.md

changelog-entry-drafter

Skill19 starsChanged 50 days agoMITView on GitHub

Draft and append a dated CHANGELOG.md entry from recent commits or a diff. Use when the user asks to update the changelog, log recent changes, or prepare release notes for one version; do not use for editing unrelated files or for a general commit-message writer.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 10 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never treat access to write a file as permission to write it.

    line 15
  2. Distinctive:

    Read the commit log and/or diff for that range.

    line 23
  3. Very specific:

    Check whether an entry for today's date already exists in CHANGELOG.md:

    line 29
  4. Distinctive:

    If yes, ask whether to append to it or replace it.

    line 31
  5. Distinctive:

    If no, prepare a new dated heading.

    line 32
  6. Distinctive:

    Default: one-time approval.

    line 40
  7. Distinctive:

    Never infer standing approval from a single approval of one entry.

    line 42
  8. Distinctive:

    Stop and ask rather than guessing when:

    line 50
  9. Distinctive:

    Before writing, confirm that:

    line 59
  10. Distinctive:

    every line in the draft traces to an observed commit/diff or a reported user description, labelled accordingly.

    line 61
  11. Distinctive:

    no secret, token, or credential appears in the drafted text.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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 Entry Drafter

Draft one dated CHANGELOG.md entry from recent commits or a diff, show it before writing, and write only with the user's approval.

Operating boundary

  • Touch only CHANGELOG.md. Do not edit source files, other docs, or configuration.
  • Base every drafted line on an actual commit message, diff hunk, or user-supplied description. Do not invent changes that are not evidenced.
  • Redact anything that looks like a secret, token, or credential found in commit messages or diffs; never copy it into the changelog.
  • Never treat access to write a file as permission to write it. Always get approval first (see below).

Workflow

1. Gather evidence

Identify the range to summarise: since the last tag, since the last changelog entry, a supplied commit range, or a supplied diff. State which one you used.

Read the commit log and/or diff for that range. Label each drafted line as based on an observed commit/diff, or reported if it comes from the user's own description rather than something you inspected.

2. Draft the entry

Group changes using Keep a Changelog–style categories (Added, Changed, Fixed, Removed, Security) — see references/changelog-entry-format.md. Omit categories with nothing to report. Keep each line short and user-facing; skip purely internal changes unless the user asks to include them.

Check whether an entry for today's date already exists in CHANGELOG.md:

  • If yes, ask whether to append to it or replace it. Do not silently duplicate.
  • If no, prepare a new dated heading.

3. Show the draft and get approval

Show the complete proposed entry to the user before writing anything. Do not write first and confirm after.

Approval model:

  • Default: one-time approval. Ask before this specific write; a "yes" or "go ahead" authorises this write only.
  • Session-scoped approval: only if the user's own request explicitly grants standing permission for the session (for example, "update the changelog after each change, don't ask me again this session"). When this applies, say so out loud before proceeding — e.g. "Proceeding without per-entry confirmation, as requested for this session" — and continue asking again in any later session.
  • Never infer standing approval from a single approval of one entry. Never infer it from tool-level file-write access being available.

4. Write and confirm

Append (or replace, per step 2) the approved entry at the correct position in CHANGELOG.md. Confirm what was written and where.

Stopping conditions

Stop and ask rather than guessing when:

  • the commit range or diff to summarise is ambiguous;
  • the evidence does not clearly indicate user-facing impact for a change;
  • a possible secret or credential is detected in the source material; report this to the user instead of drafting around it silently;
  • CHANGELOG.md does not exist or does not follow a recognisable format — ask whether to create one or use a different location.

Quality check

Before writing, confirm that:

  • every line in the draft traces to an observed commit/diff or a reported user description, labelled accordingly;
  • the entry was shown to the user before any write occurred;
  • the approval obtained matches what is being written (one-time vs. session-scoped, and for this write specifically);
  • no secret, token, or credential appears in the drafted text.

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.