Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to create, optimize, or structure a changelog or release notes page. Also use when the user mentions "changelog," "release notes," "what's new," "updates," "product updates," "version history," or "changelog.yourdomain.com." For sitewide page planning, use website-structure.
.claude/skills/kostja94-changelog-page-generator/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flashlowest | 95% | 21 |
| gemini-3.1-pro-preview | 100% | 2 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -8% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 141% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 53% | 0% |
Guides changelog and release notes pages. Typically at changelog.* subdomain or /changelog. Builds trust, reduces support, increases feature adoption.
When invoking: On first use, if helpful, open with 1–2 sentences on what this skill covers and why it matters, then provide the main output. On subsequent use or when the user asks to skip, go directly to the main output.
Check for project context first: If .claude/project-context.md or .cursor/project-context.md exists, read it for product and release cadence.
Identify:
| Section | Purpose | |---------|---------| | Entry | Date, version, title | | Category | New, Improved, Fixed, Deprecated (optional) | | Description | What changed, why it matters | | Link | To docs, blog, or in-app | | Media | Screenshots, GIFs, videos (optional) |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 17,566 | 12,662 | -28% | 1 | 1 | 0% | 2,824 | 2,590 | -8% | 0 | 0 | — |
case-02 | pass→pass | 6,363 | 11,436 | +80% | 1 | 1 | 0% | 1,044 | 2,516 | +141% | 0 | 0 | — |
case-03 | pass→pass | 6,805 | 7,140 | +5% | 1 | 1 | 0% | 1,111 | 1,695 | +53% | 0 | 0 | — |
case-04 | pass→pass | 9,426 | 9,856 | +5% | 1 | 1 | 0% | 1,851 | 2,277 | +23% | 0 | 0 | — |
case-05 | pass→pass | 6,147 | 5,647 | -8% | 1 | 1 | 0% | 1,148 | 1,653 | +44% | 0 | 0 | — |
case-06 | pass→pass | 6,038 | 5,756 | -5% | 1 | 1 | 0% | 1,113 | 1,577 | +42% | 0 | 0 | — |
case-07 | pass→pass | 13,182 | 9,344 | -29% | 1 | 1 | 0% | 2,118 | 2,070 | -2% | 0 | 0 | — |
case-08 | pass→pass | 13,708 | 7,648 | -44% | 1 | 1 | 0% | 2,231 | 1,735 | -22% | 0 | 0 | — |
case-09 | pass→pass | 10,845 | 6,731 | -38% | 1 | 1 | 0% | 1,805 | 1,661 | -8% | 0 | 0 | — |
case-10 | pass→pass | 9,224 | 9,662 | +5% | 1 | 1 | 0% | 1,235 | 2,003 | +62% | 0 | 0 | — |
case-11 | pass→pass | 23,291 | 12,131 | -48% | 1 | 1 | 0% | 2,387 | 2,552 | +7% | 0 | 0 | — |
case-12 | pass→pass | 15,979 | 15,256 | -5% | 1 | 1 | 0% | 2,579 | 2,917 | +13% | 0 | 0 | — |
case-13 | pass→pass | 23,583 | 14,495 | -39% | 1 | 1 | 0% | 3,809 | 2,878 | -24% | 0 | 0 | — |
case-14 | pass→pass | 14,515 | 14,440 | -1% | 1 | 1 | 0% | 2,346 | 2,986 | +27% | 0 | 0 | — |
case-15 | pass→pass | 15,591 | 13,252 | -15% | 1 | 1 | 0% | 2,360 | 2,638 | +12% | 0 | 0 | — |
case-16 | pass→pass | 6,878 | 8,186 | +19% | 1 | 1 | 0% | 957 | 1,855 | +94% | 0 | 0 | — |
case-17 | pass→pass | 14,623 | 15,907 | +9% | 1 | 1 | 0% | 2,192 | 2,927 | +34% | 0 | 0 | — |
case-18 | fail→pass | 15,250 | 13,540 | -11% | 1 | 1 | 0% | 2,150 | 2,468 | +15% | 0 | 0 | — |
case-19 | fail→pass | 10,332 | 2,954 | -71% | 1 | 1 | 0% | 1,720 | 1,102 | -36% | 0 | 0 | — |
case-20 | pass→pass | 16,359 | 15,990 | -2% | 1 | 1 | 0% | 2,671 | 3,113 | +17% | 0 | 0 | — |
case-21 | pass→pass | 15,612 | 13,165 | -16% | 1 | 1 | 0% | 3,272 | 3,196 | -2% | 0 | 0 | — |
case-22 | pass→pass | 17,175 | 10,803 | -37% | 1 | 1 | 0% | 3,688 | 2,866 | -22% | 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. The headline lift of +9 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.