sprint-orchestrator: Command for Claude Code

Workstream Agent Initializer

Run the sprint resume script to load workstream information, then provide agent instructions.

Step 1: Load Workstream Info

First, run the resume command to see workstream details:

pnpm sprint:resume {arg1}

Step 2: Navigate to Worktree

IMPORTANT: You must navigate to the worktree directory. Run this command:

cd ../worktrees/{arg1}

Verify you're in the correct location:

pwd  # Should show: <project-root>/../worktrees/{arg1}
git branch  # Should show: feature/{arg1}-workstream

Note: Worktrees are created in a sibling worktrees/ directory relative to the project root. The exact path depends on where you cloned the repository.

Step 3: Understand Your Role

You are a WORKSTREAM AGENT for this Sprint. You are NOT the main orchestrator.

Your Mission: Complete all tasks assigned to the {arg1} workstream.

Your Workflow

1. Check your tasks: Look at the sprint:resume output for your assigned TASK-XXX items
2. Implement each task sequentially: Follow TDD workflow (Red → Green → Refactor)
3. Run quality gates after each task:

   # Run your project's quality gates (configured in .claude/quality-gates.json)
   pnpm sprint:quality-gates
   
   # Examples for different project types:
   # Python: pytest && mypy . && ruff check .
   # Rust: cargo test && cargo check && cargo clippy
   # Go: go test ./... && go vet ./...
   # JavaScript/TypeScript: pnpm test run && pnpm type-check && pnpm lint && pnpm build
   

4. Commit after each completed task
5. When ALL tasks complete: Run pnpm sprint:complete {arg1} (runs quality gates automatically if enabled)

Important Boundaries

DO:

• ✅ Work ONLY on tasks assigned to the {arg1} workstream
• ✅ Commit after each completed task with proper format
• ✅ Run quality gates before each commit (use pnpm sprint:quality-gates or your project's commands)
• ✅ Follow TDD: write tests first, then implementation