.claude/skills/a-team/SKILL.md
Entry point that spawns the Team Architect coordinator to run the full team design workflow
npx skillsauth add chemistrywow31/A-Team A-TeamInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Launch the Team Architect coordinator to run the complete team design workflow (Phase 1-6). Use this skill as the standard entry point for all team design requests.
Use when the user wants to design, create, or build a new multi-agent team structure.
When this skill is invoked, spawn the Team Architect agent to handle the entire workflow:
team-architect agent via the Agent tool with subagent_type Team ArchitectUse the Agent tool with these parameters:
subagent_type: Team Architectmodel: opusprompt: Include the user's original request and any arguments. If the user provided no details, instruct the Team Architect to begin with Phase 1 Discovery (requirements interview).If the user provides arguments after /A-Team, pass them as context:
/A-Team 英文教學內容團隊
→ Spawn Team Architect with prompt: "Design a team for: 英文教學內容團隊. Begin with Phase 1 Discovery."
/A-Team --restructure teams/existing-team
→ Spawn Team Architect with prompt: "Restructure the existing team at teams/existing-team. Run Phase 7."
/A-Team
→ Spawn Team Architect with prompt: "The user wants to design a new team. Begin with Phase 1 Discovery — start the requirements interview."
User: /A-Team 自動化測試團隊
Action: Spawn Team Architect → "Design a team for: 自動化測試團隊. Begin with Phase 1 Discovery."
User: /A-Team --restructure teams/english-teaching-content
Action: Spawn Team Architect → "Restructure the existing team at teams/english-teaching-content. Run Phase 7 (Team Restructuring)."
User: /A-Team
Action: Spawn Team Architect → "The user wants to design a new team. Begin with Phase 1 Discovery — start the requirements interview."
data-ai
Evaluate whether a proposed agent graph supports safe handoffs and useful parallel work
testing
Extract precise team requirements through staged, single-direction questioning
content-media
Search external sources for reusable skills before designing a new one
data-ai
Create or improve a skill by using the system skill when available and the legacy local bundle as fallback