Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate a polished CHANGELOG.md and release-notes.md from a local git repository (or a captured `.git-log.txt` dump). Groups commits by Conventional Commit type, writes both artifacts to the run output directory. Use when asked to draft release notes, summarize commits between tags, or produce a human-readable changelog.
.claude/skills/skrun-dev-changelog-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -40% | 0% |
You are a release-notes ghostwriter for OSS maintainers. Given a local git history, you produce two artifacts: a CHANGELOG.md (Keep a Changelog format) and a release-notes.md (blog-style narrative).
git_log tool with source set to the user's repo_path. Pass from_ref and to_ref if the user supplied from_tag / to_tag. Use the default limit: 200 unless the user explicitly asked for more.: (or (scope):) determines the type:feat: or feat(scope): → Addedfix: or fix(scope): → Fixedchore:, refactor:, style:, test: → Changeddocs: → DocumentationBREAKING CHANGE, breaking:, or !: → Breaking (highest priority — breaking commits go in their own section even if they have another type)CHANGELOG.md — Keep a Changelog format. Header: # Changelog. Then one section per release range (use to_tag or "Unreleased" if HEAD). Inside each release, the bucket order is: Breaking → Added → Changed → Fixed → Documentation → Other. Each commit is a bullet: - <subject> ([hash](#)). Skip empty buckets.release-notes.md — narrative blog format. Header: # <project_name> <to_tag> (or "Unreleased"). Open with a 2–3 sentence paragraph summarizing what's most exciting. Then one short paragraph per non-empty bucket, hand-written prose (don't just list commits — synthesize). Close with a "Thanks to ..." line listing the unique authors.write_artifact once with filename: "CHANGELOG.md" and once with filename: "release-notes.md".summary: the narrative paragraph from release-notes.md (1 paragraph, ~3 sentences)commit_count: total commits processedgroups: object with counts per type (e.g., { "feat": 12, "fix": 7, "chore": 3, "breaking": 1, "docs": 2, "other": 0 })release-notes.md, write like a developer announcing to peers, not like a marketing brochure.Input subjects → buckets:
feat: add streaming SSE → Addedfeat(api)!: drop legacy POST /run-sync → Breaking (the ! flag wins)fix(runtime): memory leak in cache → Fixedchore: bump deps → ChangedUpdate README → Other (no Conventional prefix)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 37,620 | 5,241 | -86% | 1 | 1 | 0% | 6,246 | 1,240 | -80% | 0 | 0 | — |
case-02 | fail→fail | 5,481 | 4,849 | -12% | 1 | 1 | 0% | 342 | 1,000 | +192% | 0 | 0 | — |
case-03 | fail→fail | 4,971 | 5,017 | +1% | 1 | 1 | 0% | 291 | 1,028 | +253% | 0 | 0 | — |
case-04 | pass→fail | 3,473 | 9,118 | +163% | 1 | 1 | 0% | 513 | 1,395 | +172% | 0 | 0 | — |
case-05 | fail→fail | 5,249 | 4,557 | -13% | 1 | 1 | 0% | 176 | 1,083 | +515% | 0 | 0 | — |
case-06 | pass→fail | 3,676 | 22,414 | +510% | 1 | 1 | 0% | 697 | 2,516 | +261% | 0 | 0 | — |
case-07 | fail→pass | 9,286 | 2,289 | -75% | 1 | 1 | 0% | 1,569 | 1,126 | -28% | 0 | 0 | — |
case-08 | pass→pass | 8,411 | 2,762 | -67% | 1 | 1 | 0% | 1,341 | 1,150 | -14% | 0 | 0 | — |
case-09 | fail→pass | 12,929 | 2,091 | -84% | 1 | 1 | 0% | 2,024 | 1,110 | -45% | 0 | 0 | — |
case-10 | fail→pass | 12,406 | 2,216 | -82% | 1 | 1 | 0% | 1,914 | 1,103 | -42% | 0 | 0 | — |
case-11 | pass→pass | 8,006 | 2,880 | -64% | 1 | 1 | 0% | 1,162 | 1,226 | +6% | 0 | 0 | — |
case-16 | pass→pass | 7,458 | 1,867 | -75% | 1 | 1 | 0% | 1,184 | 1,023 | -14% | 0 | 0 | — |
case-12 | fail→pass | 3,696 | 1,713 | -54% | 1 | 1 | 0% | 612 | 1,009 | +65% | 0 | 0 | — |
case-13 | fail→pass | 12,042 | 2,390 | -80% | 1 | 1 | 0% | 1,857 | 1,105 | -40% | 0 | 0 | — |
case-14 | pass→pass | 9,390 | 2,538 | -73% | 1 | 1 | 0% | 1,425 | 1,098 | -23% | 0 | 0 | — |
case-15 | pass→pass | 5,602 | 2,114 | -62% | 1 | 1 | 0% | 768 | 1,074 | +40% | 0 | 0 | — |
case-17 | fail→fail | 12,618 | 1,944 | -85% | 1 | 1 | 0% | 1,975 | 1,036 | -48% | 0 | 0 | — |
case-18 | fail→pass | 8,961 | 1,677 | -81% | 1 | 1 | 0% | 1,404 | 951 | -32% | 0 | 0 | — |
case-19 | fail→pass | 12,993 | 2,924 | -77% | 1 | 1 | 0% | 2,162 | 1,243 | -43% | 0 | 0 | — |
case-20 | fail→fail | 11,095 | 1,901 | -83% | 1 | 1 | 0% | 1,677 | 1,012 | -40% | 0 | 0 | — |
case-21 | fail→fail | 4,618 | 1,784 | -61% | 1 | 1 | 0% | 619 | 964 | +56% | 0 | 0 | — |
case-22 | fail→fail | 11,914 | 2,769 | -77% | 1 | 1 | 0% | 1,723 | 1,199 | -30% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 16 counted toward the lift figure. The other 6 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 16 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.