agent-teams

触发: Path C/D 的 E(执行) 阶段

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "agent-teams" with this command: npx skills add wenjunduan/rlues/wenjunduan-rlues-agent-teams

触发: Path C/D 的 E(执行) 阶段

前提

  • .claude/agents/ 目录下定义了 builder/validator/explorer 等子代理

  • plan.md 中任务已标注可并行/有依赖

编排策略

  • 读 plan.md → 识别无依赖的任务组

  • 为每组分配 builder 子代理 (background: true)

  • 每个 builder 使用 isolation: worktree 隔离

  • validator 定期检查各 builder 产出

  • explorer 负责跨模块调研

并行规则

  • 无依赖任务: 并行执行

  • 有依赖任务: 串行, 前置完成后触发

  • 冲突检测: 同文件不并行修改

子代理调用

  • Task(builder): 开发任务

  • Task(validator): 验证任务

  • Task(explorer): 调研任务

  • claude agents 查看运行状态

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

tdd

No summary provided by upstream source.

Repository SourceNeeds Review
General

context7

No summary provided by upstream source.

Repository SourceNeeds Review
Security

security-review

No summary provided by upstream source.

Repository SourceNeeds Review
General

brainstorm

No summary provided by upstream source.

Repository SourceNeeds Review