team-pilot

Set up and operate TeamPilot from zero on a clean machine. Use when cloning the TeamPilot repository, installing dependencies, starting the multi-agent runtime, driving missions in auto/manual mode, replaying runs in web UI, and troubleshooting first-run issues.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "team-pilot" with this command: npx skills add moolean/team-pilot

Bootstrap and run TeamPilot end-to-end.

Zero-to-run setup (clean environment)

  1. Ensure prerequisites:
    • git --version
    • node --version (recommend Node 20+)
    • npm --version
  2. Clone repository:
    • git clone <TEAM_PILOT_REPO_URL> team-pilot
    • cd team-pilot
  3. Install dependencies:
    • npm install
  4. Start service:
    • npm run up
  5. Open web UI:
    • http://localhost:3333

If EADDRINUSE appears, stop the process occupying port 3333 or start with another port:

  • PORT=3334 npm run up
  • then open http://localhost:3334

First mission (recommended flow)

  1. In Mission Control, select a template.
  2. Set execution mode:
    • auto for smoke tests
    • manual for deterministic demos and controlled outputs
  3. Fill mission goal and launch.
  4. Confirm graph starts changing (agent lane → task lane → result lane).

Manual mode operation standard

For each task, update in two steps:

  1. task_started (status=running, progress around 20-60)
  2. task_done (status=done, progress=100, complete output)

Endpoint:

  • POST /api/tasks/:taskId/update

Suggested payload fields:

  • status, progress, output, score, eventType, assignee

Replay workflow

  1. Switch to Replay mode.
  2. Load a run with frames > 1.
  3. Scrub slider to inspect each frame.
  4. Open node detail to verify input/output at each step.

Minimal API checklist

  • POST /api/missions (create mission)
  • GET /api/state (live state)
  • POST /api/tasks/:taskId/update (manual updates)
  • GET /api/runs (list traces)
  • GET /api/runs/:missionId (load replay trace)
  • POST /api/missions/:missionId/mode (auto/manual switch)

Troubleshooting

  • UI opens but no updates:
    • verify server log has no crash
    • verify WebSocket /ws is connected
  • Replay seems flat:
    • ensure selected run has frames > 1
    • ensure task updates were posted to /api/tasks/:taskId/update
  • Mission stuck queued:
    • check dependency chain and assignee spelling
    • in manual mode, ensure leader is sending task updates

Output quality bar (content pipelines)

  • PLAN: goal, structure, acceptance criteria
  • RESEARCH: sources, extracted facts, citations
  • VERIFY: confidence, risks, uncertainty
  • BUILD: complete final draft
  • DELIVER: publish-ready summary + readiness note

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.

Automation

技能编辑器

编辑、完善或审查 AgentSkills。当需要创建新技能、对现有 SKILL.md 进行修改、清理/审计/整理技能文件时激活此技能。触发词:编辑技能, skill 注意事项, metadata 检查, 完善技能, 清理技能, 审计技能, skill 规范, 编写 skill, 新建技能

Registry SourceRecently Updated
Automation

全闭环管道

全闭环自动化管道 — Hunter→Skill Factory→Orchestrator→Dashboard→Profit。将Phase 1-3所有组件串联为自动运行的超级管道。核心能力:(1) 一键全流程 (2) 定时自动运行 (3) 异常自愈 (4) 利润报告

Registry SourceRecently Updated
Automation

智美人AI实战课

《智美人AI Agent实战课》配套技能——从0到1搭建AI Agent的完整课程体系。第01节:AI Agent基础概念+工具链搭建。课程内容含数字人讲解视频、实战代码、课后练习。覆盖:OpenClaw配置、技能安装、MCP工具、多Agent协同、变现实战。

Registry SourceRecently Updated
Automation

利润优化引擎

利润优化引擎 — 订单管理/计价/结算模拟。核心能力:(1) 订单管理 (2) 计价模型 (3) 成本追踪 (4) 利润计算

Registry SourceRecently Updated