feature-executionlisted
Install: claude install-skill stepanenkoviktor0110-boop/ai-dev-methodology
# Feature Execution
> **CRITICAL:** NEVER generate multiple artifacts without stopping. After EACH artifact: list controversial points, explain simply, WAIT for user decision. Only then proceed.
Team lead orchestrates feature delivery. You are a dispatcher: spawn agents, track progress, commit code, escalate issues. Delegate all code reading, diff analysis, and report review to spawned agents. Your only inputs are status messages from teammates ("Task complete") and escalation requests.
Before starting, read [quick-ref-feature-execution.md](../quick-learning/references/quick-ref-feature-execution.md) — top reasoning patterns for this skill (if file exists and non-empty).
## Phase 1: Initialization
0. Check `work/{feature}/logs/checkpoint.yml`:
- `last_completed_wave > 0` → this is a resume after context compaction.
Read checkpoint, then read `work/{feature}/decisions.md` to confirm what was actually completed.
For tasks in the resumed wave: if a task has a decisions.md entry, it completed — update its
frontmatter to `done` and skip it. Only re-execute tasks without a decisions.md entry.
Check if `~/.claude/teams/{team_name}/config.json` exists: if yes, team is alive; if no,
recreate via TeamCreate. Skip to Phase 2 starting from `next_wave`.
Read session-plan.md to determine which session the next_wave belongs to. Update current_session accordingly.
Report to user: "Resuming from wave {N} (session {S}). Waves 1-{N-1} completed."
-