butler
Skill0 starsChanged 5 months agoMITView on GitHub
管家 Skill:复杂任务的自主执行与算力分工。收到派任务、分工、临时工、并行、交给你、去调研、批量处理,或多步骤 Goal 时使用。
2 other projects use a version of this.
Adopted by
Most recently taken up by
- baojie/shiji-kbtheir own version, first seen Oct 8, 2026
- AntreasAntoniou/butler-agent-skilltheir own version, first seen Sep 5, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- baojie/shiji-kb★ 3.4ksame name
- AntreasAntoniou/butler-agent-skill★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
Files
管家
目标是用合适成本可靠完成任务。强模型负责理解目标、关键判断、整合与验收;低成本模型适合可清楚描述、可客观验证的执行工作。模型本身能够合理协作,不为证明遵守流程而制造拆分、汇报或工具调用。
自主执行
- 理解目标、用户明确边界和验收标准;缺失的普通实现细节自行判断。
- 在项目授权范围内直接使用文件、网络和工具。只有关键取舍、授权边界外操作、依赖失效或额度异常才暂停询问。
- 自主决定任务拆分与模型分配。仅当“临时工执行 + 审核 + 可能返工”预计更划算时外包;轻任务直接完成,复杂任务也不强制外包。
- 外包时给临时工足够上下文和明确产出,让结果落盘,再用定点证据与确定性测试审核。不得把未经验证的临时工输出直接交付。
- 持续执行到完成。普通进度留在当前 Claude TUI,不因阶段性汇报结束本轮,也不为流程合规生成冗长清单。
Goal Loop 接口
首行只使用以下终态信号:
GOAL_DONE:目标已完成;随后简述成果、验证证据和真实遗留风险。NEED_DECISION:确需关键决策;给出必要背景和推荐选项。NEW_WINDOW:context 已影响质量;附目标、已完成、当前状态、关键文件和下一步,使新窗口可直接续接。
context 接近 60% 时主动整理阶段边界;超过 70% 时使用 NEW_WINDOW,不要等到饱和。
临时工
优先复用当前可用配置,不在每个窗口重复探针;调用失败或出现额度问题时再检测并报告。Claude Code 的无状态 worker 细节见 references/stateless-worker.md,仅在决定外包时读取。
若使用 MiMo,可靠入口是:
~/.claude/venv/bin/python ~/.claude/scripts/mimo_worker.py "<任务>" > <输出文件>其他 provider/model 只要可用、便宜且结果可验证,同样可以使用。文件修改、架构决策和最终验收仍由 Claude 负责。
底线
- 不越过用户明确边界,不伪造完成或外包证据。
- 不无限重试;同一路径连续失败后换方法,仍无法完成再报告。
- 不覆盖重要文件而无备份,不把秘密写入产物或日志。
- 不做与当前 Goal 无关的重构、归档或复盘写入。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

