agent-teams

Overview of agent teams plugin with available commands, skills, and quickstart guidance

Aggregate Build profile retrospective data to surface spec quality threshold calibration recommendations

Manage project pipelines — create, list, status, next stage, skip, and archive multi-stage projects that wrap related team spawns

Unified build command — spawn a feature development or debugging team with adaptive sizing, discovery interview, and verbosity control

Unified create command — spawn a design, brainstorming, or productivity team with adaptive sizing, discovery interview, and verbosity control

Unified think command — spawn a research, planning, or review team with adaptive sizing, discovery interview, and verbosity control

AI-assisted agent team designer that helps users create custom team configurations based on their specific project needs. Can apply deep behavioral personas to teammates for structured methodology, scoring criteria, and interaction patterns. Use this agent when the user asks to "design a team", "create a custom team", "help me plan a team", "what team do I need", or when the pre-built blueprints don't fit their use case. <example> Context: User wants a team for a task that doesn't match existing blueprints user: "I need a team to migrate our database from MySQL to PostgreSQL" assistant: "I'll use the team-architect agent to design a custom team for your migration." <commentary> Database migration spans multiple concerns (schema, data, app code, testing) — needs custom team design. </commentary> </example> <example> Context: User isn't sure what team structure they need user: "What kind of team should I use for this project?" assistant: "I'll use the team-architect agent to analyze your needs and recommend a team structure." <commentary> User needs guidance on team composition. Agent will ask questions to understand the task. </commentary> </example> <example> Context: User wants to customize an existing blueprint user: "I want the feature team but with a DevOps teammate instead of a tester" assistant: "I'll use the team-architect agent to customize the feature team blueprint." <commentary> Modifying existing blueprint. Agent will adapt the template to the user's needs. </commentary> </example>

This skill should be used when the user asks to review a completed team's process, run an after-action review, or evaluate team effectiveness. Use this skill when the user says "after action review", "AAR", "process review", "team retrospective", "review team process", "how did the team work", or at the prompt after a team completes its work. Distinct from `/evaluate-spawn` (output quality). AAR evaluates team process. Both can run in the same session.

This skill should be used when the user asks about Claude Code agent teams, wants to understand team-based development workflows, or needs guidance on when and how to use agent teams. Use this skill when the user asks to "create a team", "use agent teams", "coordinate agents", "parallel development", "team workflow", or says "how do agent teams work". This skill provides an overview of agent teams, when to use them vs subagents, and routes to focused skills for team blueprints and coordination patterns.

This skill should be used when the user asks to evaluate a completed team spawn session, capture session learnings, or review how a spawn went. Use this skill when the user says "evaluate spawn", "evaluate-spawn", "how did the spawn go", "capture learnings", "review spawn session", "spawn retrospective", or at the soft prompt after a team completes its work. This is a voluntary post-spawn evaluation with profile-based questions tailored to all spawn types (build, think, create). Build profile asks up to 4 questions (3 explicit + 1 conditional gate bypass); Think and Create profiles ask up to 3 questions (2 explicit + 1 gate bypass). Hard cap includes auto-derived scenario coverage for Build.

This skill should be used when the user wants pre-designed agent team configurations for common application development phases. Use this skill when the user asks for a "research team", "feature development team", "code review team", "debug team", "design team", "planning team", "roadmap team", "team blueprint", "team template", or says "spawn a team for [development phase]". Provides 8 ready-to-use team blueprints: Research & Discovery, Feature Development, Code Review & QA, Debugging & Investigation, Frontend Design, Planning & Roadmapping, Productivity Systems, and Brainstorming & Ideation.

This skill should be used when the user needs guidance on managing an active agent team, coordinating tasks between teammates, handling team communication, or understanding team lifecycle. Use this skill when the user asks about "task management", "team communication", "delegate mode", "plan approval", "shutdown teammates", "team messaging", or says "how do I manage my team". Covers task management, messaging patterns, plan approval workflow, delegate mode, display modes, and graceful shutdown.

This skill should be used when the user asks about personas for agent teams, wants to understand how to apply deep behavioral profiles to teammates, or needs guidance on the productivity loop. Use this skill when the user asks about "personas", "behavioral profiles", "productivity loop", "auditor persona", "architect persona", "analyst persona", "refiner persona", "compounder persona", or says "what personas are available" or "how do personas work". Provides 8 reusable persona definitions that add deep methodology, scoring criteria, and interaction patterns to agent team teammates.

3 hooks across 2 events