spawn

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/hub:spawn — Launch Parallel Agents

/hub:spawn — 启动并行Agent

Spawn N subagents that work on the same task in parallel, each in an isolated git worktree.

在独立的git worktree中生成N个并行处理同一任务的子Agent。

Usage

使用方法

/hub:spawn                                    # Spawn agents for the latest session
/hub:spawn 20260317-143022                    # Spawn agents for a specific session
/hub:spawn --template optimizer               # Use optimizer template for dispatch prompts
/hub:spawn --template refactorer              # Use refactorer template

/hub:spawn                                    # 为最新会话生成Agent
/hub:spawn 20260317-143022                    # 为指定会话生成Agent
/hub:spawn --template optimizer               # 使用optimizer模板生成调度提示词
/hub:spawn --template refactorer              # 使用refactorer模板

Templates

模板

When

--template <name>

is provided, use the dispatch prompt from

references/agent-templates.md

instead of the default prompt below. Available templates:

Template	Pattern	Use Case
`optimizer`	Edit → eval → keep/discard → repeat x10	Performance, latency, size reduction
`refactorer`	Restructure → test → iterate until green	Code quality, tech debt
`test-writer`	Write tests → measure coverage → repeat	Test coverage gaps
`bug-fixer`	Reproduce → diagnose → fix → verify	Bug fix with competing approaches

When using a template, replace all

{variables}

with values from the session config. Assign each agent a different strategy appropriate to the template and task — diverse strategies maximize the value of parallel exploration.

当使用

--template <name>

参数时，将使用

references/agent-templates.md

中的调度提示词，而非下方的默认提示词。可用模板：

模板	模式	使用场景
`optimizer`	编辑 → 评估 → 保留/舍弃 → 重复10次	性能优化、延迟降低、体积缩减
`refactorer`	重构 → 测试 → 迭代直至通过	代码质量提升、技术债务处理
`test-writer`	编写测试 → 度量覆盖率 → 重复	测试覆盖率缺口填补
`bug-fixer`	复现 → 诊断 → 修复 → 验证	采用多种方案修复Bug

使用模板时，需将所有

{variables}

替换为会话配置中的值。为每个Agent分配不同的策略，使其适配模板与任务——多样化策略能最大化并行探索的价值。

What It Does

工作流程

Load session config from

.agenthub/sessions/{session-id}/config.yaml

For each agent 1..N:
- Write task assignment to
```
.agenthub/board/dispatch/
```
- Build agent prompt with task, constraints, and board write instructions
Launch ALL agents in a single message with multiple Agent tool calls:

Agent(
  prompt: "You are agent-{i} in hub session {session-id}.

Your task: {task}

Read your full assignment at .agenthub/board/dispatch/{seq}-agent-{i}.md

Instructions:
1. Work in your worktree — make changes, run tests, iterate
2. Commit all changes with descriptive messages
3. Write your result summary to .agenthub/board/results/agent-{i}-result.md
   Include: approach taken, files changed, metric if available, confidence level
4. Exit when done

Constraints:
- Do NOT read or modify other agents' work
- Do NOT access .agenthub/board/results/ for other agents
- Commit early and often with descriptive messages
- If you hit a dead end, commit what you have and explain in your result",
  isolation: "worktree"
)

Update session state to
```
running
```
via:

bash

python {skill_path}/scripts/session_manager.py --update {session-id} --state running

从

.agenthub/sessions/{session-id}/config.yaml

加载会话配置

为每个Agent（1..N）执行以下操作：
- 将任务分配写入
```
.agenthub/board/dispatch/
```
- 结合任务、约束条件和看板写入指令构建Agent提示词
在单条消息中启动所有Agent，包含多个Agent工具调用：

Agent(
  prompt: "你是hub会话{session-id}中的agent-{i}。

你的任务：{task}

请在.agenthub/board/dispatch/{seq}-agent-{i}.md查看完整任务分配

指令：
1. 在你的worktree中工作——修改代码、运行测试、迭代优化
2. 提交所有更改并添加描述性提交信息
3. 将结果摘要写入.agenthub/board/results/agent-{i}-result.md
   内容需包含：采用的方案、修改的文件、可用指标、置信度
4. 完成后退出

约束：
- 请勿读取或修改其他Agent的工作内容
- 请勿访问其他Agent的.agenthub/board/results/目录
- 尽早并频繁提交，提交信息需具有描述性
- 若陷入瓶颈，提交已完成的工作并在结果中说明原因",
  isolation: "worktree"
)

通过以下命令将会话状态更新为
```
running
```
：

bash

python {skill_path}/scripts/session_manager.py --update {session-id} --state running

Critical Rules

关键规则

All agents in ONE message — spawn all Agent tool calls simultaneously for true parallelism
isolation: "worktree" is mandatory — each agent needs its own filesystem
Never modify session config after spawn — agents rely on stable configuration
Each agent gets a unique board post — dispatch posts are numbered sequentially

所有Agent在单条消息中启动——同时生成所有Agent工具调用以实现真正的并行
isolation: "worktree"为必填项——每个Agent需要独立的文件系统
生成后请勿修改会话配置——Agent依赖稳定的配置信息
每个Agent对应唯一的看板帖子——调度帖子按顺序编号

After Spawn

生成后操作

Tell the user:

{N} agents launched in parallel
Each working in an isolated worktree
Monitor with
```
/hub:status
```
Evaluate when done with
```
/hub:eval
```

告知用户：

已并行启动{N}个Agent
每个Agent在独立的worktree中工作
使用
```
/hub:status
```
监控状态
完成后使用
```
/hub:eval
```
进行评估