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 messages-ops 用の日本語翻訳が必要です
.claude/skills/affaan-m-messages-ops/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 128% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 127% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 215% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 181% | 0% |
当任务涉及实时消息检索时使用此功能：iMessage、私信、近期一次性验证码，或后续操作前的线程检查。
这不属于邮件处理。如果主要操作界面是邮箱，请使用 email-ops。
在相关情况下，将这些 ECC 原生技能纳入工作流程：
email-ops 当消息任务实际上是邮箱操作时connections-optimizer 当私信线程属于对外网络工作时lead-intelligence 当实时线程应指导目标定位或预热路径外联时knowledge-ops 当线程内容需要捕获到持久化上下文中时在执行任何操作之前，先确定：
如果任务可能转为对外跟进：
对于一次性验证码：
返回：
text来源 - 消息界面 - 发送者 / 线程 / 服务 结果 - 消息摘要或代码 - 时间窗口 状态 - 已读 / 已找到代码 / 受阻 / 等待回复草稿
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 13,769 | 16,683 | +21% | 1 | 1 | 0% | 360 | 1,597 | +344% | 0 | 0 | — |
case-02 | pass→pass | 2,844 | 7,217 | +154% | 1 | 1 | 0% | 388 | 1,815 | +368% | 0 | 0 | — |
case-03 | pass→fail | 4,827 | 7,483 | +55% | 1 | 1 | 0% | 663 | 1,865 | +181% | 0 | 0 | — |
case-04 | pass→pass | 10,922 | 6,728 | -38% | 1 | 1 | 0% | 1,687 | 1,718 | +2% | 0 | 0 | — |
case-05 | pass→pass | 11,679 | 9,883 | -15% | 1 | 1 | 0% | 1,760 | 2,066 | +17% | 0 | 0 | — |
case-06 | fail→pass | 5,098 | 5,769 | +13% | 1 | 1 | 0% | 681 | 1,551 | +128% | 0 | 0 | — |
case-07 | pass→pass | 5,373 | 7,318 | +36% | 1 | 1 | 0% | 762 | 1,717 | +125% | 0 | 0 | — |
case-08 | pass→pass | 2,798 | 5,660 | +102% | 1 | 1 | 0% | 381 | 1,542 | +305% | 0 | 0 | — |
case-09 | pass→pass | 2,971 | 5,345 | +80% | 1 | 1 | 0% | 495 | 1,569 | +217% | 0 | 0 | — |
case-10 | fail→pass | 8,214 | 4,203 | -49% | 1 | 1 | 0% | 1,315 | 1,362 | +4% | 0 | 0 | — |
case-11 | pass→pass | 3,308 | 6,699 | +103% | 1 | 1 | 0% | 469 | 1,692 | +261% | 0 | 0 | — |
case-12 | fail→pass | 4,208 | 5,570 | +32% | 1 | 1 | 0% | 654 | 1,482 | +127% | 0 | 0 | — |
case-13 | pass→pass | 4,122 | 4,924 | +19% | 1 | 1 | 0% | 554 | 1,423 | +157% | 0 | 0 | — |
case-18 | pass→pass | 5,994 | 7,180 | +20% | 1 | 1 | 0% | 899 | 1,795 | +100% | 0 | 0 | — |
case-14 | pass→pass | 5,097 | 6,049 | +19% | 1 | 1 | 0% | 721 | 1,559 | +116% | 0 | 0 | — |
case-15 | pass→pass | 6,281 | 9,414 | +50% | 1 | 1 | 0% | 376 | 2,160 | +474% | 0 | 0 | — |
case-16 | pass→pass | 3,041 | 8,758 | +188% | 1 | 1 | 0% | 453 | 1,987 | +339% | 0 | 0 | — |
case-17 | pass→pass | 6,739 | 9,511 | +41% | 1 | 1 | 0% | 422 | 1,546 | +266% | 0 | 0 | — |
case-19 | fail→pass | 4,470 | 7,523 | +68% | 1 | 1 | 0% | 575 | 1,811 | +215% | 0 | 0 | — |
case-20 | pass→pass | 3,382 | 9,285 | +175% | 1 | 1 | 0% | 421 | 2,059 | +389% | 0 | 0 | — |
case-21 | pass→pass | 4,077 | 6,453 | +58% | 1 | 1 | 0% | 551 | 1,564 | +184% | 0 | 0 | — |
case-22 | pass→pass | 2,802 | 6,675 | +138% | 1 | 1 | 0% | 371 | 1,629 | +339% | 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 +14 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.