team-assemble

Team Assemble

Analyze a task, dynamically design the right expert agents, and orchestrate them as a team using Claude Code's agent teams feature.

Prerequisites

Agent teams must be enabled. See references/enable-agent-teams.md for setup instructions.

When to Use

• Complex tasks that decompose into 2+ independent subtasks
• Work where role separation is clear (e.g., research + implementation + validation)
• Tasks that benefit from parallel execution

Do NOT use for: single-file edits, simple questions, purely sequential work

---

Core Principles

• Model 3-tier — choose by role purpose (details: references/agents.md)
  • opus — strategy/judgment (scouts, complex execution)
  • sonnet — standard execution/validation (worker, qa, support)
  • haiku — exploration/writing (researcher, editor)
• No fixed catalog — agents are designed dynamically per task
• Example bank — references/agents.md provides reference examples (not mandatory)

---

Workflow

Phase 1 → Phase 2 → Phase 3 → Phase 4 → Phase 5 → Phase 6
Task       Codebase   Integrate  Execute   Validate   Complete
Analysis   Scouts     & Confirm                       & Cleanup
                                           ↕ FAIL → support fix (max 3x)

---

Phase 1: Task Analysis

Analyze the user's request and identify relevant areas of the codebase:

1. Examine project structure, CLAUDE.md files, and README files
2. Identify which parts of the codebase are relevant to the task
3. Determine if codebase scouting (Phase 2) would help, or if the task is straightforward enough to skip to Phase 3

Get user approval via AskUserQuestion:

I've analyzed your task and identified the following areas of interest:

- [x] src/auth/ — Authentication module (needs refactoring)
- [x] tests/auth/ — Corresponding tests
- [ ] src/api/ — Not directly affected

I'll scout these areas to design an optimal team.

Options: "Looks good, proceed" / "I'd like to adjust the scope"