subagent-orchestrator-dsh
turambarr/subagent-orchestrator-dsh
Experiment-orchestration SOP skill for long-running experiments. A top-level model drafts a decomposition plan — subtasks, the model for each, an acceptance figure with tolerance, and dependency order — and dispatches only after the user confirms that plan; each subtask then goes to a lightweight executor subagent as a self-contained prompt, and its report is confined to measured facts and the delta from the agreed figure, with judgement and decisions staying at the top level. Ships one plan template and five task templates (gather, execute, consistency check, read-only review, mechanical transform), acceptance gates split into a task layer and an experiment layer, and three rules that stop internal-consistency evidence from being presented as independent or external validation.
Installazione
dsh plugin --profile web add github:turambarr/subagent-orchestrator-dshREADME
subagent-orchestrator-dsh
实验编排 SOP 技能（DeepSeek Harness Skill）：顶层模型（pro）控制实验设计与验收，智能调用轻量子智能体（flash）执行确定性子任务。
这是什么
一个把「pro 控制 / flash 执行」的实验分工写死的可复用技能，解决长周期实验里"每一步该拆给谁、怎么下指令、怎么验收"的问题。
核心机制：
- 唯一硬规则：flash 能接 = 过程确定 + 派工前能写下「数字 + 容差」。
- 分解计划：每实验一次，pro 起草、用户确认后才派工（顶层控制显式化、可审查）。
- 自包含 prompt 模板：采集 / 执行 / 校验 / 审查 / 转换 五类，flash 零上下文也能干。
- 两级验收 + 四级证据门 + 三条红线：低级证据不许冒充高级证据。
结构
skills/experiment-orchestration/
├── SKILL.md # 主指令：核心循环 + 硬规则 + 模型映射
├── prompt-templates/
│ ├── plan.md # 分解计划模板
│ ├── fact-gather.md # 采集
│ ├── execute-verify.md # 执行
│ ├── consistency-check.md # 一致性校验
│ ├── read-only-review.md # 只读审查
│ └── mechanical-transform.md # 机械转换
└── acceptance/
└── gates.md # 两级验收 + 四级门 + 三条红线
安装
dsh plugin --profile <profile> add subagent-orchestrator-dsh
领域扩展（可选）
本技能不内置领域专属内容。需要接入特定领域时，新增 domain/<name>/appendix.md，核心机制不动。
License
MIT