bmad-quick-dev

Implements code changes through a spec-driven plan, implement, and adversarial review workflow.

Updated Jul 9, 2026
One-click install
npx skills add https://github.com/unbakedcosmos/thumbnailer --skill bmad-quick-dev-unbakedcosmos
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bmad-quick-dev
Source: https://github.com/unbakedcosmos/thumbnailer/tree/main/.claude/skills/bmad-quick-dev
Command: npx skills add https://github.com/unbakedcosmos/thumbnailer --skill bmad-quick-dev-unbakedcosmos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a feature request or bug fix into reviewed, working code often skips planning and review discipline, producing unreviewed diffs and scope creep. This Skill enforces a structured pipeline that clarifies intent, writes a bounded spec, implements it, and runs adversarial review before presenting the result. ## Core Features & Use Cases - Spec-driven routing: Clarifies user intent, checks scope (single goal, 900–1600 tokens), and routes work to either a one-shot path or a full plan-code-review cycle with human approval checkpoints. - Adversarial review loop: Launches Blind Hunter and Edge Case Hunter reviewers on the diff, classifies findings (intent_gap, bad_spec, patch, defer, reject), and loops back to re-plan or re-implement when needed. - Epic context compilation: Distills planning artifacts (PRD, architecture, UX) into a focused per-epic context file and syncs story status with sprint-status.yaml. - Use Case: Ask it to add a retry button to a queue row; it drafts a spec with tasks and acceptance criteria, gets your approval, implements the change, reviews the diff, commits, and opens a clickable Suggested Review Order in VS Code. ## Quick Start Ask the agent to implement a specific feature or bug fix, for example: implement a numbered-copy naming scheme for overwritten contact sheet outputs.

Frequently Asked Questions about bmad-quick-dev

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I implement a feature with a spec-driven AI workflow?▼

Invoke the skill with your intent; it clarifies requirements, writes a spec file with tasks and acceptance criteria, waits for your approval, then implements and reviews the change. You approve at checkpoints before code is written.

What is the difference between the one-shot and plan-code-review routes?▼

One-shot handles zero-blast-radius changes with clear intent, implementing directly with a single review pass. Plan-code-review covers everything else with a full spec, human approval checkpoint, implementation, and dual adversarial review.

How does the adversarial code review loop work?▼

Two reviewers, Blind Hunter and Edge Case Hunter, inspect the diff without prior context. Findings are classified as intent_gap, bad_spec, patch, defer, or reject; serious findings trigger spec amendment and re-implementation, capped at five loop iterations.

Does this workflow require subagents to run?▼

No. Subagents are preferred for exploration, implementation, and review, but each step has an inline fallback. When subagents are unavailable, review prompts are written to files for you to run in separate sessions.

Can it resume interrupted or previously drafted specs?▼

Yes. It scans the implementation artifacts directory for specs with draft, ready-for-dev, in-progress, or in-review status and resumes at the matching step. It also restores epic context and prior story continuity automatically.