tjdrhs90/rn-launch-harness/rn-harness-spec

Claude Code and agent skill by tjdrhs90/rn-launch-harness

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Phase 2.5 — Break PRD into file-level task checklists under docs/specs/ for progress tracking.

Install

npx vskill@latest install tjdrhs90/rn-launch-harness/rn-harness-spec

SKILL.md

rn-harness-spec — Phase 2.5: Spec Planning

Breaks down the PRD into file-level task checklists under docs/specs/, organized by feature and implementation phase. Sits between Plan (Phase 2) and Design (Phase 3).

Trigger

Called by the orchestrator after Phase 2 (Plan) completes.

Input

  • docs/harness/plans/YYYY-MM-DD-prd.md (PRD from Phase 2)
  • docs/harness/config.md

Process

Step 1: Extract Feature List

Read the PRD and extract the full list of features to implement. Map each feature to the FSD module structure defined in the PRD (features/, entities/, widgets/, shared/).

Sort features by priority:

  • P0 features first (MVP required)
  • P1 features next (important)
  • P2 features last (future roadmap)
Step 2: Create Feature Spec Directories

For each feature, create a numbered directory under docs/specs/:

docs/specs/
├── 01-auth/
├── 02-home/
├── 03-profile/
├── ...
└── README.md

The NN prefix reflects implementation priority order (01, 02, 03...).

Step 3: Generate Phase Files

Inside each feature directory, create phase markdown files:

docs/specs/01-auth/
├── phase1-mvp.md # Core MVP functionality
├── phase2-enhancement.md # Enhancements and extensions
└── phase3-polish.md # Optimization and polish (if needed)

Phase breakdown rules:

  • Phase 1 (MVP): Minimum viable implementation — types, API, store, core UI, screen, barrel export
  • Phase 2 (Enhancement): Extended features — validation, error states, loading states, edge cases
  • Phase 3 (Polish): Optimization — animations, accessibility, performance tuning

P0 features should have at least Phase 1 and Phase 2. P1 features may only have Phase 1. P2 features are documented but not given phase files (noted in README as future scope).

Step 4: Write Tasks in Checkbox Format

Each phase file must follow this exact structure:

markdown
---
feature: auth
phase: 1
title: MVP - Authentication
status: not-started
created: YYYY-MM-DD
updated: YYYY-MM-DD
---
# Phase 1: MVP - Authentication
## Tasks
### Entity Setup
-[ ] `src/entities/user/types/index.ts` — IUser interface definition
-[ ] `src/entities/user/store/index.ts` — Zustand user store
### Feature Implementation
-[ ] `src/features/auth/api/auth.api.ts` — Login/signup API calls
-[ ] `src/features/auth/hooks/useLogin.ts` — Login mutation hook
-[ ] `src/features/auth/hooks/useRegister.ts` — Signup mutation hook
-[ ] `src/features/auth/store/index.ts` — Token management store
-[ ] `src/features/auth/ui/LoginForm.tsx` — Login form component
-[ ] `src/features/auth/types/index.ts` — Auth-related types
-[ ] `src/features/auth/index.ts` — Barrel export
### Screen Implementation
-[ ] `app/(auth)/login.tsx` — Login screen
-[ ] `app/(auth)/register.tsx` — Register screen
### QA
-[ ] typecheck passes
-[ ] lint passes
-[ ] feature functions correctly

Task format rules:

  • Every task line starts with - [ ]
  • File path in backticks, followed by — (em dash) and a brief description
  • Group tasks by FSD layer: Entity Setup, Feature Implementation, Screen Implementation, Widget (if applicable), Shared (if applicable)
  • Always end each phase with a QA section
Step 5: Generate Progress Dashboard

Create docs/specs/README.md as a progress dashboard:

markdown
# Spec Dashboard
> Generated: YYYY-MM-DD
> Source PRD: docs/harness/plans/YYYY-MM-DD-prd.md
## Progress Overview
| # | Feature | Phase 1 | Phase 2 | Phase 3 | Status |
|---|---------|---------|---------|---------|--------|
| 01 | auth | 0/12 | 0/8 | - | not-started |
| 02 | home | 0/9 | 0/5 | - | not-started |
| 03 | profile | 0/7 | - | - | not-started |
## Feature Details
### 01-auth
-Phase 1: MVP - Authentication (12 tasks)
-Phase 2: Enhancement - Auth polish (8 tasks)
### 02-home
-Phase 1: MVP - Home feed (9 tasks)
-Phase 2: Enhancement - Feed optimization (5 tasks)
...
## Future Scope (P2)
-[ ] Feature X — deferred to post-launch
-[ ] Feature Y — deferred to post-launch
Step 6: User Confirmation

Present the spec structure summary via AskUserQuestion:

  • Total number of features and tasks
  • Phase breakdown per feature
  • Estimated scope
  • Ask for adjustments before proceeding

Spec Frontmatter Schema

yaml
---
feature: {feature-name} # Matches directory name (without NN prefix)
phase: {1|2|3} # Implementation phase
title: {phase title} # Human-readable title
status: {not-started|in-progress|completed}
created: {YYYY-MM-DD}
updated: {YYYY-MM-DD}
---

Checkbox Update Rules (Phase 5: Generator)

During Phase 5 (Generator), as the Generator implements each file:

  1. Update the corresponding task: - [ ] to - [x]
  2. Update the frontmatter updated date
  3. When all tasks in a phase are checked, set status: completed
  4. Update docs/specs/README.md dashboard counts (e.g., 3/12)
  5. Update dashboard status column based on progress:
  • not-started — 0 tasks completed
  • in-progress — at least 1 task completed
  • completed — all tasks completed

Output

docs/specs/
├── 01-{feature}/
│ ├── phase1-mvp.md
│ ├── phase2-enhancement.md
│ └── phase3-polish.md (optional)
├── 02-{feature}/
│ ├── phase1-mvp.md
│ └── phase2-enhancement.md
├── ...
└── README.md

State Update

yaml
current_phase: design
next_role: rn-harness-design

HARD GATES

  • Every task must include a file path — no vague descriptions like "implement auth"
  • File paths must follow FSD layer conventions (features/, entities/, shared/, widgets/)
  • YAML frontmatter required on every phase file
  • README.md dashboard must be generated
  • QA section required in every phase file
  • User confirmation required before proceeding to Phase 3
  • P0 features must have at minimum Phase 1 tasks defined

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/tjdrhs90/rn-launch-harness/rn-harness-spec/badge)](https://verified-skill.com/skills/tjdrhs90/rn-launch-harness/rn-harness-spec)

Something wrong with this skill or its page? Tell us.