Skip to content

xinetzone/SpecWeave / .agents/skills/agent-task-handoff

agent-task-handoff

Verified

当用户要把一件有外部副作用的事交给 AI Agent(浏览器自动化、手机助手、桌面 Agent、外包或实习生)执行,或出现『AI 说做了却没生效』『同一任务越跑越贵』『第一次让 AI 碰陌生项目』时,必须使用此技能。五模式编排:P1 定路线(底座先行)→ P2 建地图(只讲不改)→ P3 写任务书(做什么/条件/停点/回报/红线)→ P5 控成本(步数与停点前移)→ P4 排障(分链定位);排障时反向从 P4 回到 P3。另含四张补充卡(E1 四问成技 / E2 证据先行 / E3 单产出拆分 / E4 双闸门),覆盖把重复流程写成技能、失败先取证、多段编排与不可逆动作能力层兜底。产出可验收的委托契约、步数台账与排障记录;不替代需求澄清,不替用户执行不可逆动作。

FileKindSize
SKILL.mdskill md33.7 KB
evals/evals.jsonconfig3.6 KB
references/step-ledger.mdmarkdown2.2 KB
references/task-brief-template.mdmarkdown3.6 KB
references/triage-worksheet.mdmarkdown2.7 KB

evals/evals.json

View on GitHub
{
  "skill_name": "agent-task-handoff",
  "evals": [
    {
      "id": 1,
      "prompt": "帮我让手机助手在淘宝买一台 14 寸轻薄本,预算 5000 以内,别太重的,看着合适就下单吧。",
      "expected_output": "识别为 T2 强信号并进入 P3:指出「看着合适就下单」既缺可判定条件又缺停点;产出四要素任务书(做什么/条件数字化/停点停在结算前/回报字段/红线不代填支付信息),并在下发前给出预览等确认,不直接执行下单。",
      "files": []
    },
    {
      "id": 2,
      "prompt": "这个浏览器自动化脚本跑起来之后,模型说它点了按钮,但页面根本没变化,已经重装两次了,你帮我看看怎么回事。",
      "expected_output": "进入 P4:给出理解链与执行链两条最小指令,按「只理解不执行」分支只查三项(执行器配置/输入模态支持/权限或无障碍),强调以外部可观察状态为证据,一次只改一项并做交叉回归;明确指出不应继续重装。",
      "files": []
    },
    {
      "id": 3,
      "prompt": "我想让 AI 第一次读我们这个老项目,然后直接帮我加一个导出 CSV 的功能。",
      "expected_output": "先执行 P2:给三问(关键模块/请求链路/落点)+ 末尾写死「不要修改任何文件」,以工作区零改动为验收;再进入 P3 写任务书把「加导出功能」转成可验收条件与停点,说明顺序不可颠倒。",
      "files": []
    },
    {
      "id": 4,
      "prompt": "我要做一个每周选题→写稿→排版→发布的内容流水线,交给几个 AI 助手分工做,发布前必须我确认。",
      "expected_output": "先走 E3:按可验收产出切段(选题/写稿/排版/发布),每段一张六字段交接单,发布设为人工关口,先跑最费劲的写稿段,调度者按验收项自核、不过不传;再对发布这类不可逆动作上 E4:规则层写具体项确认、能力层要求确认凭据缺失即拒绝,并做反证测试。说明不按人数拆、也不把确认只写在提示词里。",
      "files": []
    },
    {
      "id": 5,
      "prompt": "我想让手机助手以后写邮件默认都用我固定的简洁语气,同事说这得改它的程序代码,你帮我安排一下怎么改。",
      "expected_output": "进入 P1 判层:这是「偏好」而非能力缺失,应落记忆层(或技能层规则),先用可声明资产表达、不下沉程序层,并指出「改程序代码」是反模式;产出落点表与三元组台账(改了什么/怎么证明生效含多久生效/怎么退回),一次只改一个落点并在真实环境验证。若该产品是三层均不可配置的黑盒 SaaS,或改动必须跨三层同时生效,则声明 P1 不适用并回到底座评估。",
      "files": []
    },
    {
      "id": 6,
      "prompt": "我刚改了助手的一条技能规则,顺手还调了两个配置项,现在打算换个说法考考它,看看改得行不行。",
      "expected_output": "触发 P3 试改验纪律并指出两处违规:一次改三处导致无法归因(要求回退为单点改动)、换说法复测会混淆输入变化与规则变化(要求用与基线完全相同的同一句输入原样复测);给出完整五步:改前留基线→只改一处→发布(技能 lint+push 或程序构建+安装,注明即时生效还是需重装)→同句复测→对照回退;说明 lint 通过不等于行为正确、改完不复测不得进入下一需求;并指出首次搭建无基线或无法重复触发的场景不适用本循环。",
      "files": []
    }
  ]
}
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 470kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall