zero-window-taskbook-drafting
Skill3 starsChanged 8 days agoApache-2.0View on GitHub
甲类大型只有命中宪法 0.1 回退线任一条、需要双窗时,才用本 skill 起草或核对 0 号书;单窗用 short-book-drafting。按短书骨架写五块,把轻闸骨架交给独立执行员窗,核清白名单、判据、审计与偏离项。
In this repository25 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- claude-session-observabilitySkill
- closeout-report-draftingSkill
- context-guard-hooksSkill
- contract-and-test-infraSkill
- doc-pointer-checkSkill
- engineering-pitfalls-checklistSkill
- first-window-requirementsSkill
- frontend-integration-preflightSkill
- hive-dispatchSkill
- legacy-project-onboardingSkill
- memory-index-foldingSkill
- memory-writingSkill
- merge-and-deploySkill
- open-source-preflightSkill
- pointer-handoffSkill
- repo-scale-reportingSkill
- shared-checkout-safetySkill
- short-book-draftingSkill
- subtraction-reviewSkill
- ticket-desk-usageSkill
- top-audit-dispatchSkill
- verify-agent-claimsSkill
- window-handoff-draftingSkill
The file
Files
这个 skill 是干什么的(人话)
这个 skill 帮你只在甲类大型命中宪法 0.1 回退线任一条、需要双窗时写 0 号书,单窗用 short-book-drafting。它要配合 md-workflow 全套用,照短书骨架写五块,再把轻闸工序交给独立执行员窗。不用它,容易把默认单窗误写成双窗,或沿用旧六段书漏掉已定口径、判据和白名单。
分类
分类:需要 md-workflow 全套
用前填这几个名字
- {{拍板人}}:唯一有最终裁定权的人的称谓;例:产品负责人
- {{远端主支}}:远端主干分支的完整引用名,任务书里的底数都对它量;例:origin/main
- {{工单台命令}}:工单台命令行入口,第 0 步四件、免判卷标记都用它;例:
python -X utf8 core/t.py;有python用python,没有用py -X utf8 core/t.py兜底 - {{工作树根目录}}:一单一工作树统一放置的目录,扫盘子代理的 md 也落在它下面;例:../_trees
- {{顶级模型}}:顶级档模型名,主窗本体用,不派子代理;例:按模型对照表填本项目实值
- {{困难模型}}:困难档模型名,复杂施工、完整审计与起草用;例:按模型对照表填本项目实值
- {{日常模型}}:日常档模型名,普通施工、轻审与登记用;例:按模型对照表填本项目实值
- {{轻量模型}}:轻量档模型名,扫盘读件、抄数与固定脚本用,不审码、不审图;例:按模型对照表填本项目实值
什么时候用
- 甲类大型命中宪法 0.1 回退线任一条,确需单开独立执行员窗(双窗)。
- 双窗的 0 号书要起草、修改或在发出前核对五块、判据、白名单与偏离项。
- 单窗不写 0 号书,改用 short-book-drafting;不能只因工单大就另开窗。
步骤
- 先核双窗条件:甲类大型的 0 号书只在双窗(命中宪法 0.1 回退线任一条)时写;单窗用 short-book-drafting。
- 读本目录
references/taskbook-steps.md,核清交接边界与远端主干底数。 - 照
desk/模板/短书骨架.md写五块:出题、已定口径、判据与改动面、底数件路径、并线去向与审计等口径。 - 把
desk/模板/轻闸骨架.md交给独立执行员窗;施工编排由执行员写进度件第一笔,短书不代写。 - 按 references 的自检核五块、白名单、免判卷、独立审计、产出上限与偏离项,从工单台取开窗指令,不手写。
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.

