subagent-driven-developmentlisted

<!-- Adapted from superpowers (https://github.com/obra/superpowers), MIT (c) Jesse Vincent. --> # Subagent-Driven Development Execute a plan by dispatching a fresh subagent per task, with two-stage review after each: spec compliance first, then code quality. **Why subagents:** you delegate to specialized agents with isolated context. By precisely crafting their instructions you keep them focused and preserve your own context for coordination. They never inherit your session history — you construct exactly what they need. **Core principle:** Fresh subagent per task + two-stage review (spec then quality) = high quality, fast iteration. **Continuous execution:** Do NOT pause to check in between tasks. Execute all tasks. Only stop for: an unresolvable BLOCKED status, genuine ambiguity, or completion. "Should I continue?" prompts waste the user's time — they asked you to execute the plan. ## Fits in the pipeline This is the default execution engine for **Stage 6 (Build, `/build`)**. It pairs with `dispatching-parallel-agents` (for independent domains across worktrees), `using-git-worktrees` (isolation), and `requesting-code-review` (the reviewer subagents). Each task's reviews feed naturally into the **Stage 7 per-feature QA gate**. Priority: **user > skills > default**; `_shared/vibegod-principles.md` apply (subagents follow TDD and surgical-change rules). ## When to Use ```dot digraph when_to_use { "Have implementation plan?" [shape=diamond]; "Tasks mostly indep