A command that creates a separate work thread for a selected role, such as product manager, frontend developer, backend developer, tester, DevOps engineer, or technical lead. It then starts that role's task using the matching project instructions.

Use it to start a role-specific task with a short title and optional description.

It organizes work by role and keeps each task in its own thread with a title, description, and role tags.

start-task — Command by peterfei

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/peterfei/ai-agent-team/start-task

Clone the repo

git clone --depth 1 https://github.com/peterfei/ai-agent-team

Made for: Claude Code.

Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,857 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00013	$0.01857
Opus 5	$0.00006	$0.00928
Sonnet 5	$0.00003	$0.00371
Haiku 4.5	$0.00001	$0.00186

Measured 3d ago against content hash d47a020b67a9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

start-task scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/start-task.md · 262 lines

How it starts

The opening of the file, as written. The whole thing — 262 lines — stays where its author put it; the contents beside it link to each section on GitHub.

统一任务启动命令

此命令为任何 AI Agent 角色创建独立的工作线程。

使用方法

/start-task <角色> "任务标题" [--desc "描述"]

支持的角色

角色代码	角色名称	标签
`pm`	产品经理	product, pm
`fe`	前端开发	frontend, fe
`be`	后端开发	backend, be
`qa`	测试工程师	qa, testing
`ops`	DevOps工程师	devops, ops
`tl`	技术负责人	tech-lead, tl

执行流程

第1步：解析参数并创建线程

<角色>: 从支持列表中识别角色
"任务标题": 提取任务标题作为线程标题
--desc: 可选的详细描述

调用 mcp__thread-manager__create_thread：

{
  title: "[角色名称] - [任务标题]",
  description: 描述（如果提供）,
  tags: 角色对应的标签,
  switchTo: true
}

第2步：简洁通知

只输出一行：

✨ 已创建[角色]线程："[标题]" (ID: [前8位])

第3步：立即开始工作 ⭐ 最重要

不要停止！立即切换到对应角色并开始工作：

根据角色读取对应的 agent prompt
- pm → .claude/agents/product_manager.md
- be → .claude/agents/backend_dev.md
- fe → .claude/agents/frontend_dev.md
- qa → .claude/agents/qa_engineer.md
- ops → .claude/agents/devops_engineer.md
- tl → .claude/agents/tech_leader.md
以该角色的身份和能力工作
输出完整的工作成果（需求文档/代码/测试等）
就像用户直接调用了 /[角色] [任务] 一样

使用示例

产品经理任务

用户: /start-task pm "设计用户认证系统"

Claude:
✨ 已创建产品线程："产品经理 - 设计用户认证系统" (ID: abc12345)

---

## 产品需求分析 - 用户认证系统

### 执行摘要
...

[立即开始完整的产品分析工作]

后端开发任务

用户: /start-task be "实现JWT认证API" --desc "支持access token和refresh token"

Claude:
✨ 已创建后端线程："后端开发 - 实现JWT认证API" (ID: def67890)

---

## 后端技术方案 - JWT认证API

### 技术选型
...

[立即开始完整的后端开发工作]

前端开发任务

用户: /start-task fe "开发登录表单组件"

Claude:
✨ 已创建前端线程："前端开发 - 开发登录表单组件" (ID: ghi24680)

---

## 前端组件设计 - 登录表单

### 组件结构
...

[立即开始完整的前端开发工作]

工作流示例

完整产品开发流程

# 第1步：产品经理创建需求线程
用户: /start-task pm "用户认证系统需求分析"
Claude: [创建线程并返回启动命令]
用户: exit
$ clt abc123
用户在新会话: /pm "开始进行用户认证系统的需求分析"
产品经理: [在独立线程中进行需求分析...]

# 第2步：后端开发创建实现线程
用户: exit  # 退出产品经理线程
$ claude  # 回到主会话
用户: /start-task be "实现JWT认证API"
Claude: [创建新线程]
用户: exit
$ clt def456
用户在新会话: /be "根据产品需求实现JWT认证API"
后端开发: [在独立线程中开发...]

# 第3步：前端开发创建UI线程
用户: /start-task fe "开发登录UI组件"
Claude: [创建新线程]
用户: exit
$ clt ghi789
用户在新会话: /fe "开发与后端API对接的登录表单"
前端开发: [在独立线程中开发...]

# 第4步：QA测试
用户: /start-task qa "测试用户认证功能"
Claude: [创建新线程]
用户: exit
$ clt jkl012
用户在新会话: /qa "执行用户认证的完整测试"
QA工程师: [在独立线程中测试...]

Read the full file on GitHub · 262 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

3d ago First seen · 262 lines · 13 tokens per session scan A d47a020b67a9

Subscribe to this mod's changes

start-task is a command published in the GitHub repository peterfei/ai-agent-team (428 stars, last pushed 2mo ago), licensed MIT. It adds 13 tokens to every session and 1,857 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

24886 Building agents ai ai-agents claude-code claude-code-skill drawnotes

start-task