Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →自动生成 CHANGELOG，基于 git 提交历史和 pipeline 产物信息，遵循 Conventional Commits 和 Keep a Changelog 规范
.claude/skills/echovic-devops-changelog-generation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 105% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 62% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 34% | 0% |
在部署成功完成后自动生成或追加 CHANGELOG，记录本次发布的所有变更。适用于：
bash # 获取从上次 tag 到 HEAD 的所有提交 git log $(git describe --tags --abbrev=0 2>/dev/null || git rev-list --max-parents=0 HEAD)..HEAD --pretty=format:"%H|%s|%an|%ai"
type(scope): description.boss/<feature>/prd.md → 功能描述和用户价值.boss/<feature>/deploy-report.md → 部署环境和版本信息.boss/<feature>/tasks.md → 完成的任务列表package.json 中的 versionYYYY.MM.DD按 Conventional Commits 规范分类：
| 类型 | CHANGELOG 分类 | 说明 | |------|---------------|------| | feat | Added | 新增功能 | | fix | Fixed | 修复问题 | | perf | Performance | 性能优化 | | refactor | Changed | 重构（非功能变更） | | docs | Documentation | 文档更新 | | style | _(不记录)_ | 代码格式 | | test | _(不记录)_ | 测试相关 | | chore | _(不记录)_ | 构建/工具变更 | | BREAKING CHANGE | ⚠️ Breaking Changes | 破坏性变更（始终置顶） |
对于非 Conventional Commits 格式的提交：
markdown## [版本号] - YYYY-MM-DD ### ⚠️ Breaking Changes - 破坏性变更描述 ([commit-hash]) ### Added - 新功能描述（来自 PRD 的用户价值说明） ([commit-hash]) ### Changed - 变更描述 ([commit-hash]) ### Fixed - 修复描述 ([commit-hash]) ### Performance - 优化描述 ([commit-hash])
两种输出模式：
.boss/<feature>/changelog.mdCHANGELOG.md，将新版本内容追加到文件顶部（在 # Changelog 标题之后）追加逻辑：
读取现有 CHANGELOG.md
→ 找到第一个 ## [version] 行
→ 在其前面插入新版本内容
→ 写回文件如果项目无 CHANGELOG.md，则创建包含标准头部的新文件：
markdown# Changelog All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/). ## [版本号] - 日期 ...
.boss/<feature>/changelog.md 必须生成CHANGELOG.md，同步更新| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,766 | 14,285 | +21% | 1 | 1 | 0% | 1,697 | 3,471 | +105% | 0 | 0 | — |
case-02 | fail→fail | 19,285 | 5,605 | -71% | 1 | 1 | 0% | 3,027 | 1,824 | -40% | 0 | 0 | — |
case-03 | fail→pass | 9,869 | 9,614 | -3% | 1 | 1 | 0% | 1,542 | 2,498 | +62% | 0 | 0 | — |
case-04 | pass→pass | 11,939 | 12,575 | +5% | 1 | 1 | 0% | 1,977 | 3,068 | +55% | 0 | 0 | — |
case-05 | pass→pass | 18,215 | 14,074 | -23% | 1 | 1 | 0% | 3,087 | 3,485 | +13% | 0 | 0 | — |
case-06 | pass→pass | 21,902 | 16,263 | -26% | 1 | 1 | 0% | 3,856 | 4,366 | +13% | 0 | 0 | — |
case-07 | fail→pass | 14,911 | 12,267 | -18% | 1 | 1 | 0% | 2,424 | 2,988 | +23% | 0 | 0 | — |
case-08 | fail→pass | 17,245 | 11,845 | -31% | 1 | 1 | 0% | 2,597 | 2,903 | +12% | 0 | 0 | — |
case-09 | fail→pass | 7,118 | 2,994 | -58% | 1 | 1 | 0% | 1,072 | 1,436 | +34% | 0 | 0 | — |
case-10 | pass→pass | 8,308 | 2,842 | -66% | 1 | 1 | 0% | 1,303 | 1,441 | +11% | 0 | 0 | — |
case-11 | pass→pass | 7,408 | 3,264 | -56% | 1 | 1 | 0% | 1,248 | 1,516 | +21% | 0 | 0 | — |
case-12 | fail→pass | 13,767 | 6,350 | -54% | 1 | 1 | 0% | 2,017 | 1,955 | -3% | 0 | 0 | — |
case-13 | pass→pass | 15,243 | 12,091 | -21% | 1 | 1 | 0% | 2,377 | 2,889 | +22% | 0 | 0 | — |
case-14 | fail→pass | 17,821 | 6,391 | -64% | 1 | 1 | 0% | 2,880 | 2,070 | -28% | 0 | 0 | — |
case-15 | pass→pass | 12,625 | 5,849 | -54% | 1 | 1 | 0% | 1,925 | 1,891 | -2% | 0 | 0 | — |
case-16 | fail→pass | 19,817 | 12,888 | -35% | 1 | 1 | 0% | 3,403 | 3,282 | -4% | 0 | 0 | — |
case-17 | pass→pass | 14,649 | 9,032 | -38% | 1 | 1 | 0% | 2,212 | 2,383 | +8% | 0 | 0 | — |
case-18 | pass→pass | 16,507 | 11,166 | -32% | 1 | 1 | 0% | 2,682 | 2,755 | +3% | 0 | 0 | — |
case-19 | pass→pass | 11,985 | 6,030 | -50% | 1 | 1 | 0% | 1,953 | 1,963 | +1% | 0 | 0 | — |
case-20 | pass→pass | 10,791 | 4,393 | -59% | 1 | 1 | 0% | 1,874 | 1,726 | -8% | 0 | 0 | — |
case-21 | pass→pass | 16,130 | 12,113 | -25% | 1 | 1 | 0% | 2,265 | 2,944 | +30% | 0 | 0 | — |
case-22 | fail→pass | 10,393 | 2,428 | -77% | 1 | 1 | 0% | 1,638 | 1,406 | -14% | 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 +41 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.