Menu
Autonomous MVP builder — brainstorm, scaffold, build, and track web app prototypes
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Conducts structured research on a topic with web search — a Standard path for speed, a Deep path that adds discovery and claim verification. Use when the user asks to research a topic, investigate a technology, compare approaches, or gather information for a decision. Produces canonical markdown plus a portable HTML microsite in _claude/research/.
Close out a completed /dr-plan and ship it — verify all plan checkboxes, backfill the retro, move the plan to completed/, then commit, push, and open a GitHub PR populated from the plan summary. Invoked explicitly as /dr-ship.
Initializes or updates a project with the project-management plugin structure. Creates _claude/ directories and a versioned CLAUDE.md on fresh projects, verifies and updates outdated plugin-managed sections on existing projects, or appends plugin sections to an existing CLAUDE.md. Use when setting up the plugin in a new project or when plugin template sections have been updated.
Create or refine a detailed implementation plan file in `_claude/plans/` with per-phase acceptance criteria, verification gates, and safe refinement. Supports four modes: CREATE, REFINE, SUMMARY (PR summary, optional push to PR), and QUESTION RESOLUTION (interactive Q&A). Use when the user writes `/dr-plan` anywhere in their message, or explicitly asks to create, refine, or summarize a plan file under `_claude/plans/`. Do NOT use for general planning discussion, brainstorming, outlines, or task lists not destined for a file in `_claude/plans/`.
Creates or refines a Product Requirement Document (PRD) under _claude/prd/ with a structured discovery phase, adaptive sections based on feature type, and safe refinement with backup, diff preview, and linked-plan detection. Use when the user writes `/dr-prd` anywhere in their message. Do NOT use for general discussion about drafting or updating a PRD without the explicit `/dr-prd` token. Supports traditional features and AI/LLM features (adds eval rubrics, model constraints, prompt specs, performance budgets, guardrails).
This skill should be used when the user is building, modifying, or reviewing React components or applications. It provides React 19.x development patterns and APIs that should be preferred over legacy approaches. Covers Actions, useActionState, useFormStatus, useOptimistic, use() API, Form Actions, ref as prop, Context as provider, document metadata, Server Components, Server Actions, prerender(), Activity component, useEffectEvent, cacheSignal(), and Partial Pre-rendering.
| name | mvp |
| description | Autonomous MVP builder — brainstorm, scaffold, build, and track web app prototypes |
| argument-hint | <start|build|status|summary> |
| disable-model-invocation | true |
| effort | high |
| allowed-tools | Read Write Edit Grep Glob Bash(*) Agent AskUserQuestion WebSearch WebFetch TaskCreate TaskUpdate TaskGet TaskList |
Autonomous skill that brainstorms an app idea with the user, scaffolds the project, and builds a working prototype using parallel AI agents.
Effort mode: high (set by /mvp for best results)
Override with /effort medium if preferred.
Detect the mode from the user's arguments and delegate to the appropriate reference file.
Arguments: $ARGUMENTS
Route to the correct mode:
If arguments are empty, whitespace-only, or start with "start":
${CLAUDE_SKILL_DIR}/references/start.md and follow those instructionsIf arguments start with "build":
${CLAUDE_SKILL_DIR}/references/build.md and follow those instructionsIf arguments start with "status":
${CLAUDE_SKILL_DIR}/references/status.md and follow those instructionsIf arguments start with "summary":
${CLAUDE_SKILL_DIR}/references/summary.md and follow those instructionsIf no recognized mode, show usage help:
MVP Builder — Autonomous web app prototype builder
Usage:
/mvp start Brainstorm an idea, choose tech stack, scaffold project
/mvp build Build the prototype (resumes from saved state)
/mvp status View current progress dashboard
/mvp summary Generate HTML analytics page
Get started:
/mvp start
All modes inherit these conventions. Mode-specific reference files build on top of them.
All MVP state is persisted in .mvp/ in the current working directory:
| File/Folder | Purpose |
|---|---|
.mvp/brainstorm.md | Source-of-truth: vision, scope, task tracking, agent log |
.mvp/state.json | Machine-readable state for orchestration and analytics |
.mvp/agent-logs/ | Individual agent run reports (one per agent) |
.mvp/research/ | Research artifacts gathered during brainstorming |
.mvp/resources/ | Downloaded assets (images, icons, etc.) |
JavaScript Stack:
Elixir Stack:
All timestamps use ISO 8601 format: YYYY-MM-DDTHH:MM:SS (local time). Use the current date from the conversation context — NEVER hardcode dates.
mvp: [description]Subagents MUST return results as structured JSON:
{
"agentId": 1,
"status": "success|partial|failed",
"taskCompleted": "description of what was done",
"filesModified": ["path/to/file1"],
"filesCreated": ["path/to/new-file"],
"dependenciesNeeded": ["package-name"],
"processesStarted": [{"pid": 12345, "command": "npm run dev"}],
"issues": ["any problems encountered"],
"notes": "additional context"
}
state.json tracks resource locks to prevent agent conflicts:
| Lock | Protects |
|---|---|
migrations | Database schema changes — one agent at a time |
design | Global CSS, theme, design system files |
dependencies | package.json / mix.exs modifications (main agent only) |
The main agent (not subagents) ALWAYS handles:
.mvp/state.json and .mvp/brainstorm.mdUse TaskCreate and TaskUpdate to provide real-time visual progress in the Claude Code UI. This complements the .mvp/state.json system which persists across sessions.
Route to the correct mode based on the arguments above.