buildlisted

# specclaw build **First, run** `specclaw-ensure-init .specclaw` — idempotently creates `.specclaw/` if it doesn't exist (silent if already initialized; auto-inits using the current directory's basename as the project name). Execute the planned tasks. ## Step 0 — Validate Run `specclaw-validate-change .specclaw <change> build`. If it fails, report missing prerequisites and stop. ## Step 1 — Setup ```bash specclaw-build setup .specclaw <change> ``` Returns JSON config with `parallel_tasks`, `models.coding`, `git.strategy`, `notifications.channel`. Capture this — you'll use `parallel_tasks` and `model` throughout the build. **Worktree strategy:** when `git.strategy: worktree-per-change`, setup creates an isolated worktree at `.specclaw/worktrees/<change>/`. Use the `worktree_path` from the JSON as the working directory when spawning coding agents. Send a **build started** notification: ``` 🦞 **Build Started** **Change:** <change> **Branch:** specclaw/<change> **Tasks:** <total_count> across <wave_count> waves ``` ## Step 2 — Parse tasks ```bash specclaw-parse-tasks --status pending .specclaw/changes/<change>/tasks.md ``` Outputs JSON: `[{"id":"T1","title":"...","wave":1,"depends":[],"files":[...],"estimate":"small"}, ...]`. **Retry:** to re-run failed tasks, parse with `--status failed`, reset each to `pending` via `specclaw-update-task-status .specclaw/changes/<change>/tasks.md <TASK_ID> pending`, then re-parse with `--status pending`. ## Step 3 — Wave loop Fo