full-stack-feature
Meta-skill for orchestrating end-to-end feature development from persona research through deployed, tested code — coordinating skills, commands, and agents across every stage. Use when building a new feature end-to-end, coordinating frontend + backend work, or needing a structured approach from research to production.
Best use case
full-stack-feature is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Meta-skill for orchestrating end-to-end feature development from persona research through deployed, tested code — coordinating skills, commands, and agents across every stage. Use when building a new feature end-to-end, coordinating frontend + backend work, or needing a structured approach from research to production.
Teams using full-stack-feature should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/full-stack-feature/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How full-stack-feature Compares
| Feature / Agent | full-stack-feature | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Meta-skill for orchestrating end-to-end feature development from persona research through deployed, tested code — coordinating skills, commands, and agents across every stage. Use when building a new feature end-to-end, coordinating frontend + backend work, or needing a structured approach from research to production.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Full-Stack Feature Orchestration (Meta-Skill)
Coordinate the entire lifecycle of a feature — from understanding who it's for, through design, implementation, testing, and documentation. This meta-skill routes to the right skill, command, or agent at each stage and enforces stage gates so nothing ships half-baked.
## Installation
### OpenClaw / Moltbot / Clawbot
```bash
npx clawhub@latest install full-stack-feature
```
---
## When to Use
- Building a new feature end-to-end — you need the full pipeline from research to production
- Frontend + backend work together — the feature spans UI, API, and data layers
- User research should inform implementation — you want personas and specs before writing code
- Coordinating multiple skills — you're unsure which skill or command to invoke at each step
- Onboarding a new feature area — you need a structured approach, not ad hoc implementation
- Teaching a junior developer — walk through the complete feature development process
---
## Orchestration Flow
Every feature moves through eight stages. Small features skip stages (see Complexity Assessment below); large features complete all of them.
```
1. Persona Research
↓
2. Feature Specification
↓
3. UX/UI Design
↓
4. API Design
↓
5. Frontend Implementation
↓
6. Backend Implementation
↓
7. Testing
↓
8. Documentation
```
---
## Stage Details
### Stage 1 — Persona Research
Understand who the feature is for before designing anything. Define the target user, their goals, pain points, and the journey this feature supports.
**Invoke:** `/create-persona` command → `ai/commands/docs/create-persona.md`
**Skill:** `persona-docs` → `ai/skills/writing/persona-docs/SKILL.md`
**Output:** Persona document describing the target user, their context, and success criteria for this feature.
### Stage 2 — Feature Specification
Translate persona insights into a concrete feature spec — scope, acceptance criteria, edge cases, and technical constraints.
**Invoke:** `feature-specification` skill → `ai/skills/meta/feature-specification/SKILL.md`
**Output:** Feature spec with user stories, acceptance criteria, out-of-scope items, and technical notes.
### Stage 3 — UX/UI Design
Design the interface informed by the persona and spec. Choose the right visual style, layout, and interaction patterns.
**Output:** Component hierarchy, layout decisions, style selections, and responsive behavior plan.
### Stage 4 — API Design
Define the contract between frontend and backend. Design endpoints, request/response shapes, error codes, and authentication requirements.
**Command:** `/create-api-route` → `ai/commands/development/create-api-route.md`
**Output:** API contract (OpenAPI spec or typed route definitions) with validation rules and error handling.
### Stage 5 — Frontend Implementation
Build the UI components, pages, and client-side logic. Use the appropriate framework skill for the project's stack.
**Command:** `/create-component` → `ai/commands/development/create-component.md`
**Output:** Working UI components wired to the API contract, with loading/error states and responsive layout.
### Stage 6 — Backend Implementation
Build the API routes, business logic, data access layer, and background jobs.
**Command:** `/new-feature` → `ai/commands/development/new-feature.md`
**Output:** Implemented API routes, data models, migrations, and business logic passing unit tests.
### Stage 7 — Testing
Validate the feature across unit, integration, and end-to-end layers. Verify acceptance criteria from Stage 2.
**Command:** `/test-feature` → `ai/commands/development/test-feature.md`
**Output:** Passing test suite covering happy paths, edge cases, and error scenarios from the spec.
### Stage 8 — Documentation
Generate user-facing docs, API references, and internal technical documentation.
**Command:** `/generate-docs` → `ai/commands/documentation/generate-docs.md`
**Output:** Updated README, API docs, component storybook entries, and changelog.
---
## Skill Routing Table
| Stage | Primary Skill | Command | Agent |
|-------|--------------|---------|-------|
| 1. Persona Research | `persona-docs` | `/create-persona` | — |
| 2. Feature Spec | `feature-specification` | — | — |
| 3. UX/UI Design | UI/UX skill | — | — |
| 4. API Design | `api-design-principles` | `/create-api-route` | `ai/agents/api/` |
| 5. Frontend | Framework skill (Next.js, React, etc.) | `/create-component` | — |
| 6. Backend | `api-development`, `database-migration-patterns` | `/new-feature` | `ai/agents/migration/` |
| 7. Testing | `testing-workflow`, `e2e-testing-patterns` | `/test-feature` | `ai/agents/testing/` |
| 8. Documentation | — | `/generate-docs` | — |
---
## Stage Gate Checks
Each stage must pass its gate before proceeding. Gates prevent wasted work by catching gaps early.
| Gate | Required Before | Criteria | Blocking? |
|------|----------------|----------|-----------|
| Persona defined | Stage 2 | Target user identified with goals and pain points | Yes |
| Spec approved | Stage 3 | Acceptance criteria written, scope defined, edge cases listed | Yes |
| Design reviewed | Stage 4 | Component hierarchy defined, responsive plan in place | Yes |
| API contract locked | Stage 5 + 6 | Endpoints defined, request/response types agreed, error codes set | Yes |
| Frontend renders | Stage 6 | UI components display with mock data, loading/error states work | No (parallel OK) |
| Backend passes tests | Stage 7 | All API routes return expected responses, validations enforced | Yes |
| Tests pass | Stage 8 | Unit + integration + E2E tests green, acceptance criteria verified | Yes |
| Docs complete | Deploy | API documented, user-facing docs updated, changelog entry added | Yes |
---
## Vertical Slice Strategy
Start with the thinnest possible end-to-end slice, then widen.
**Phase 1 — Thin Slice**
Build one happy path through all layers: a single user action from UI click to database write and back. This proves the architecture works and gives stakeholders something to demo.
```
Example: "User can create a new project"
UI: One form with a name field and submit button
API: POST /api/projects { name: string }
DB: INSERT INTO projects (name) VALUES ($1)
Test: E2E test: fill form → submit → see project in list
```
**Phase 2 — Widen**
Add validation, error handling, edge cases, and secondary flows. Each addition follows the same vertical path — never build an entire layer in isolation.
**Phase 3 — Polish**
Loading states, optimistic updates, animations, accessibility, performance optimization, and comprehensive error messages.
---
## Complexity Assessment
Not every feature needs all eight stages. Use this table to determine which stages to include.
| Feature Size | Examples | Stages to Include | Estimated Time |
|-------------|----------|-------------------|----------------|
| **Trivial** | Rename a label, fix copy, adjust spacing | 5 only | < 1 hour |
| **Small** | Add a filter, new form field, simple toggle | 4 → 5 → 6 → 7 | 2-4 hours |
| **Medium** | New CRUD entity, dashboard widget, search feature | 2 → 3 → 4 → 5 → 6 → 7 | 1-3 days |
| **Large** | New user-facing feature area, multi-page flow | All 8 stages | 1-2 weeks |
| **Epic** | New product vertical, major redesign, platform migration | All 8 + ADR + phased rollout | 2-6 weeks |
### How to Assess Complexity
1. **Count the layers touched** — UI only (trivial), UI + API (small), UI + API + DB (medium+)
2. **Count the user flows** — one path (small), 2-3 paths (medium), many paths with branching (large)
3. **Check for unknowns** — known patterns (smaller), new integrations or unfamiliar tech (bump up one size)
4. **Consider blast radius** — isolated change (smaller), cross-cutting concern (bump up one size)
---
## Coordination Patterns
### Frontend and Backend in Parallel
Once the API contract is locked (Stage 4 gate), frontend and backend can proceed simultaneously:
- **Frontend** uses mock data matching the API contract types
- **Backend** implements against the same contract with unit tests
- **Integration** happens when both sides are ready — contract guarantees compatibility
### Handoff Points
Use the `/handoff-and-resume` command when:
- Switching between frontend and backend work
- Pausing mid-feature and resuming later
- Passing work to another developer or agent
### Progress Tracking
Use the `/progress` command to check which stage you're in and what remains.
---
## NEVER Do
1. **NEVER skip persona research for user-facing features** — building without understanding the user leads to features nobody wants
2. **NEVER start coding before the API contract is defined** — frontend and backend will diverge, causing costly rework at integration
3. **NEVER build an entire layer before connecting it end-to-end** — always use vertical slices to prove the architecture first
4. **NEVER skip stage gates to move faster** — gates exist to catch problems when they're cheap to fix, not after they've compounded
5. **NEVER treat testing as a separate phase you can cut** — tests are part of implementation, not an afterthought bolted on at the end
6. **NEVER ship without documentation** — undocumented features become maintenance burdens that slow down every future changeRelated Skills
game-changing-features
Find 10x product opportunities and high-leverage improvements. Use when the user wants strategic product thinking, mentions 10x, wants to find high-impact features, or asks what would make a product dramatically more valuable.
feature-specification
Convert persona docs into detailed feature specifications with acceptance criteria. Use when translating user needs into implementable specs, writing user stories, defining acceptance criteria, or preparing features for development.
tailwind-v4-+-shadcn/ui-stack
No description provided.
schema-markup
Add, fix, or optimize schema markup and structured data. Use when the user mentions schema markup, structured data, JSON-LD, rich snippets, schema.org, FAQ schema, product schema, review schema, or breadcrumb schema.
prompt-engineering
Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, designing production prompt templates, or building AI-powered features.
professional-communication
Write effective professional messages for software teams. Use when drafting emails, Slack/Teams messages, meeting agendas, status updates, or translating technical concepts for non-technical audiences. Triggers on email, slack, teams, message, meeting agenda, status update, stakeholder communication, escalation, jargon translation.
persona-docs
Create persona documentation for a product or codebase. Use when asked to create persona docs, document target users, define user journeys, document onboarding flows, or when starting a new product and needing to define its audience. Persona docs should be the first documentation created for any product.
mermaid-diagrams
Create software diagrams using Mermaid syntax. Use when users need to create, visualize, or document software through diagrams including class diagrams, sequence diagrams, flowcharts, ERDs, C4 architecture diagrams, state diagrams, git graphs, and other diagram types. Triggers include requests to diagram, visualize, model, map out, or show the flow of a system.
clear-writing
Write clear, concise prose for humans — documentation, READMEs, API docs, commit messages, error messages, UI text, reports, and explanations. Combines Strunk's rules for clearer prose with technical documentation patterns, structure templates, and review checklists.
brainstorming
Explore ideas before implementation through collaborative dialogue. Use before any creative work — creating features, building components, adding functionality, or modifying behavior. Turns ideas into fully formed designs and specs through structured conversation.
Article Illustrator
When the user wants to add illustrations to an article or blog post. Triggers on: "illustrate article", "add images to article", "generate illustrations", "article images", or requests to visually enhance written content. Analyzes article structure, identifies positions for visual aids, and generates illustrations using a Type x Style two-dimension approach.
subagent-driven-development
Execute implementation plans by dispatching a fresh subagent per task with two-stage review (spec compliance then code quality). Use when you have an implementation plan with mostly independent tasks and want high-quality, fast iteration within a single session.