Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Jnmetacode/Changelog — measured AI agent skill | DecimalAI

jnmetacode/changelog

SkillScore91Maintained

Generate or update a CHANGELOG.md in Keep a Changelog format from real git history. Use when cutting a release, tagging a version, or asked "what changed since…".

by jnmetacode

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Doc Ops pack

decimalai skills pull jnmetacode-changelog

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/jnmetacode-changelog/SKILL.md

Raw file: /s/jnmetacode-changelog/SKILL.md ↗ · machine-readable summary at /s/jnmetacode-changelog.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 7d ago

Yes — +5% lift+0% tokens0% turns91% of whole cases pass with skill (20/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-15	✗→✓	▲ Improved	-3%	0%
case-03	✓→✓	＝ Same ✓	-15%	0%
case-04	✓→✓	＝ Same ✓	-37%	0%
case-05	✓→✓	＝ Same ✓	-26%	0%
case-06	✓→✓	＝ Same ✓	50%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

jnmetacode / skillet

skills/changelog/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

Skill contents

SKILL.md

changelog

A changelog is for users deciding whether to upgrade, not a commit dump. Format: Keep a Changelog + semver sections.

Procedure

Find the range. Last release tag: git describe --tags --abbrev=0.

Collect commits since: git log <tag>..HEAD --oneline --no-merges.

Bucket each commit:

Added (feat) · Fixed (fix) · Changed (behavior changes, perf)
Deprecated · Removed · Security
Drop pure chore/ci/test/docs noise unless user-visible.

Rewrite each line user-first. Translate implementation into impact:

commit: fix(ingest): treat mtime regressions as edits
entry: Fixed duplicate chunks after restoring files from backup.

One line each; link the PR/issue number when the repo uses them.

Squash threads. Five commits iterating on one feature = one entry.
Pick the version bump from the buckets: breaking → major, Added → minor,

only Fixed → patch (pre-1.0: breaking → minor).

Write the section at the top of CHANGELOG.md:

markdown ## 0.2.0] - 2026-06-10 ### Added

Live re-indexing: engram watch picks up edits automatically.

### Fixed

Duplicate chunks after restoring files from backup.

Keep an ## [Unreleased] section at the very top for changes landing between

releases; releasing = renaming it and starting a fresh one.

Quality bar

Every entry answers: what can I do now / what stopped being broken? If an entry needs the source code to make sense, rewrite it.

Want to see if jnmetacode/changelog would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.