do-tasklisted

# Do Task > **CRITICAL:** NEVER generate multiple artifacts without stopping. After EACH artifact: list controversial points, explain simply, WAIT for user decision. Only then proceed. Execute a spec-driven task with validation and status tracking. Before starting, read [quick-ref-do-task.md](../quick-learning/references/quick-ref-do-task.md) — top reasoning patterns for this skill (if file exists and non-empty). ## Step 1: Read Task 1. Read task file (user provides path or task number) - If user didn't specify → ask: "Which task to execute?" 2. Verify task status is `planned` (if not → ask user before proceeding) 3. Update task frontmatter: `status: planned` → `status: in_progress` 4. Read every file listed in the task's "Context Files" section ## Step 2: Execute 1. Load each skill listed in the task (frontmatter `skills: [...]` and "Required Skills" section) - If a skill is not found → warn user, continue with remaining skills - If task has no skill (frontmatter `skills: []` or absent) → read the task, execute "What to do" and "Verification Steps" directly. For tasks with user instructions → show the instruction to user, wait for confirmation. 2. Follow loaded skill workflow 3. Git commit implementation (code + tests pass): `feat|fix|refactor: task {N} — {brief description}` 4. For each reviewer from the task's "Reviewers" section (if present): 1. Spawn subagent via Task tool (subagent_type = reviewer name, e.g. `code-reviewer`) 2. Pass: git diff of ch