Install any skill in seconds. Free to start, no credit card required.
Get Started Free →mohitagw15856/changelog-writer
Turn a list of changes, commits, or PRs into clean release notes / a changelog entry.
Passed 17 of 22 test cases with the skill, and 12 of 22 without it.
and 7 more in the full report
7 fixed10 passed either way3 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
Format a release entry for our desktop app dated August 20, 2025. Changes: - Resolved window flickering on dual monitors (#512) - Fixed memory leak when closing tabs (#518). Current version is 3.1.4. What is the next version?
| Checked | Without skill | With skill |
|---|---|---|
| Assigns version 3.1.5 as the new release version | Passed | Passed |
| Groups both changes under the Fixed section | Missed | Passed |
The instructions your model reads when the skill loads.
Raw commit logs are written for the author; a changelog is written for the user. This skill turns a pile of commits/PRs/changes into a clean release entry — grouped by type, in plain user-facing language, with breaking changes and upgrade steps surfaced first so nobody gets surprised.
Ask for these only if they aren't already provided:
Follow Keep a Changelog conventions:
⚠️ Breaking changes (only if any) — each breaking change + the exact migration step to fix it. This goes first.
Added — new features/capabilities, in user terms. Changed — changes to existing behavior. Deprecated — soon-to-be-removed features (and what to use instead). Fixed — bug fixes (what was broken, from the user's view). Security — any security-relevant fixes.
(Omit empty sections.) Each line: user-facing outcome first, with an issue/PR reference if available — not the raw commit message.
Upgrade notes (if needed) — anything to do when upgrading beyond the breaking-changes steps.
Semver note — if the version was inferred, one line on why (breaking → major, feature → minor, fix → patch).
test cases passed, vs 12 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.