Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →日本語翻訳：このファイルは project-flow-ops 用の日本語翻訳が必要です
.claude/skills/affaan-m-project-flow-ops/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 5% | 0% |
此技能将分散的 GitHub Issue、PR 和 Linear 任务整合为一条执行流程。
当问题在于协调而非编码时使用。
收集：
每个项目应归入以下状态之一：
| 状态 | 含义 | |-------|---------| | 合并 | 独立完整、符合策略、准备就绪 | | 移植/重建 | 有用的想法，但应在 ECC 内部手动重新落地 | | 关闭 | 方向错误、过时、不安全或重复 | | 搁置 | 可能有用，但当前未排期 |
仅在以下情况下创建或更新 Linear：
不要机械地镜像所有内容。
当工作活跃时：
当工作完成或被拒绝时：
返回：
text公开状态 - 议题 / 拉取请求状态 - 持续集成 / 审查状态 分类 - 合并 / 移植重建 / 关闭 / 搁置 - 一段理由说明 线性操作 - 创建 / 更新 / 无需线性项 - 项目 / 泳道（如适用） 下一步操作者行动 - 确切的下一个步骤
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,658 | 13,027 | -22% | 1 | 1 | 0% | 2,422 | 2,725 | +13% | 0 | 0 | — |
case-02 | fail→pass | 14,384 | 11,259 | -22% | 1 | 1 | 0% | 2,193 | 2,240 | +2% | 0 | 0 | — |
case-03 | fail→pass | 17,909 | 9,727 | -46% | 1 | 1 | 0% | 2,588 | 2,216 | -14% | 0 | 0 | — |
case-04 | pass→pass | 11,738 | 7,356 | -37% | 1 | 1 | 0% | 1,974 | 1,818 | -8% | 0 | 0 | — |
case-05 | pass→pass | 12,087 | 7,984 | -34% | 1 | 1 | 0% | 2,019 | 1,990 | -1% | 0 | 0 | — |
case-06 | pass→pass | 12,843 | 9,220 | -28% | 1 | 1 | 0% | 2,229 | 2,178 | -2% | 0 | 0 | — |
case-07 | fail→pass | 14,637 | 9,792 | -33% | 1 | 1 | 0% | 2,133 | 2,208 | +4% | 0 | 0 | — |
case-08 | pass→pass | 13,444 | 10,385 | -23% | 1 | 1 | 0% | 1,735 | 2,274 | +31% | 0 | 0 | — |
case-13 | pass→pass | 13,355 | 7,209 | -46% | 1 | 1 | 0% | 1,961 | 1,814 | -7% | 0 | 0 | — |
case-09 | pass→pass | 11,653 | 4,947 | -58% | 1 | 1 | 0% | 1,727 | 1,599 | -7% | 0 | 0 | — |
case-10 | fail→pass | 14,505 | 9,810 | -32% | 1 | 1 | 0% | 2,155 | 2,235 | +4% | 0 | 0 | — |
case-11 | fail→pass | 13,053 | 8,823 | -32% | 1 | 1 | 0% | 1,861 | 1,957 | +5% | 0 | 0 | — |
case-12 | pass→pass | 28,464 | 7,663 | -73% | 1 | 1 | 0% | 2,350 | 1,874 | -20% | 0 | 0 | — |
case-14 | pass→pass | 13,710 | 10,978 | -20% | 1 | 1 | 0% | 1,939 | 2,271 | +17% | 0 | 0 | — |
case-15 | pass→pass | 13,109 | 11,438 | -13% | 1 | 1 | 0% | 1,918 | 1,818 | -5% | 0 | 0 | — |
case-16 | pass→pass | 8,349 | 5,277 | -37% | 1 | 1 | 0% | 1,279 | 1,600 | +25% | 0 | 0 | — |
case-17 | pass→pass | 8,253 | 5,121 | -38% | 1 | 1 | 0% | 1,332 | 1,546 | +16% | 0 | 0 | — |
case-18 | pass→pass | 14,703 | 7,918 | -46% | 1 | 1 | 0% | 2,107 | 1,940 | -8% | 0 | 0 | — |
case-19 | fail→pass | 17,467 | 10,412 | -40% | 1 | 1 | 0% | 2,498 | 2,175 | -13% | 0 | 0 | — |
case-20 | fail→pass | 17,586 | 8,740 | -50% | 1 | 1 | 0% | 2,507 | 2,065 | -18% | 0 | 0 | — |
case-21 | pass→pass | 10,552 | 5,460 | -48% | 1 | 1 | 0% | 1,391 | 1,624 | +17% | 0 | 0 | — |
case-22 | fail→pass | 20,001 | 10,016 | -50% | 1 | 1 | 0% | 2,545 | 2,283 | -10% | 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 +36 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.