piv-speckit
🥈SilverPIV Spec-Kit is a structured framework for AI-assisted development. It combines Prime-Implement-Validate methodology with strict Test-Driven Development. Benefits developers by ensuring high code quality and reducing errors. Integrates with Claude AI for code generation and validation. Suitable for Node.js, Python, React, and Spring Boot projects.
2720Updated 3mo ago
Intermediate30min to implementautomation
Saves ~180 min per use
Quick InstallView Source
git clone https://github.com/galando/piv-speckit.gitWorks with:
Claude
Overview
About This Skill
PIV Spec-Kit is a Claude Code plugin that implements the Prime-Implement-Validate methodology for AI-assisted software development. It automatically generates structured specification artifacts (spec.md, plan.md, tasks.md, quickstart.md) and enforces strict Test-Driven Development by requiring tests to be written before implementation. The framework auto-primes context when planning features, auto-activates validation skills including TDD compliance checking, code review, and security scanning, and organizes all artifacts in a version-controlled .claude directory. Developers use /piv-speckit:plan-feature to create specs, /piv-speckit:execute to implement with TDD, and automatic validation ensures test coverage, security, and methodology compliance.
How to Use
[{"step":"Prime Phase: Start with your requirements document. Break down the feature into 3-5 core requirements and identify potential edge cases. Use the prompt template to generate the initial design analysis.","action":"Copy the prompt template into Claude, replace [PLACEHOLDERS] with your specific requirements, and run the analysis. Review the design decisions before proceeding.","tip":"For complex features, break them into smaller sub-features first. Use bullet points in your requirements to make them easier to process."},{"step":"Implement Phase: Generate the code using the validated design. Ensure all functions have JSDoc comments and follow your project's style guide.","action":"Use the implementation output as a starting point. Modify the code to fit your existing codebase structure. Add any additional helper functions needed.","tip":"For large features, implement one function at a time and validate each before proceeding to the next. Use your IDE's refactoring tools to maintain consistency."},{"step":"Validate Phase: Create comprehensive unit tests that cover all code paths, including error conditions. Aim for 100% code coverage.","action":"Use the test framework specified in your prompt (Jest, pytest, etc.). Run tests locally first, then integrate with your CI/CD pipeline.","tip":"Write tests for both happy paths and edge cases. Use mocking for external dependencies like databases or APIs. Consider property-based testing for complex logic."},{"step":"Iterate: Review the test results and code coverage. Refactor if needed, then repeat the validation phase until all tests pass.","action":"Use tools like Istanbul for coverage reports. Address any uncovered code paths by adding more tests or modifying the implementation.","tip":"Set up pre-commit hooks to run tests automatically before code is committed. This ensures no untested code enters your repository."},{"step":"Document: Add the generated documentation to your project's README or wiki. Include the design decisions and test coverage metrics.","action":"Update your project documentation with the new feature's API reference, usage examples, and test results. Include any known limitations.","tip":"For team projects, create a pull request with the complete PIV Spec-Kit output attached. This gives reviewers context about the implementation decisions."}]
Use Cases
Plan and implement new features with structured requirements and technical specs
Enforce Test-Driven Development workflows across team projects
Generate automated validation including security scanning and code review
Organize project context and specifications in version-controlled artifacts
Tags
Setup & Installation
Quick Install
No install command available. Check the GitHub repository for manual installation instructions.
Alternative Install (Git Clone)
git clone https://github.com/galando/piv-speckitRequirements
- Claude Code or compatible AI agent
- Works with: Claude
Quick Start Guide
1
Install the Skill
Copy the install command above and run it in your terminal.
2
Open Your AI Agent
Launch Claude Code, Cursor, or your preferred AI coding agent.
3
Try It Out
Use the prompt template or examples below to test the skill.
4
Customize
Adapt the skill to your specific use case and workflow.
Usage Examples
Prompt Template
Act as a PIV Spec-Kit (Prime-Implement-Validate + Test-Driven Development) assistant for [LANGUAGE/PROJECT TYPE]. Follow this process: 1) Prime: Analyze the requirements for [FEATURE] and generate a high-level design with edge cases. 2) Implement: Write the code for [FEATURE] in [LANGUAGE] with clear comments. 3) Validate: Create unit tests using [TEST_FRAMEWORK] that cover 100% of the new code. Use [CODE_STYLE_GUIDE] for consistency. Start with: 'Prime: [REQUIREMENTS]'
Example Output
### Prime Analysis for User Authentication Feature (Node.js/Express)
**Requirements:**
- Users must register with email/password (min 8 chars, valid email format)
- Passwords hashed with bcrypt (cost factor 12)
- JWT tokens expire in 24h
- Rate limiting: 5 failed attempts per IP
- Email verification required (token expires in 1h)
**Design Decisions:**
- Use Express middleware for auth validation
- Store users in PostgreSQL with `created_at` and `verified_at` timestamps
- Implement token refresh endpoint
- Edge cases: concurrent registrations, expired tokens, SQL injection attempts
### Implementation (Node.js/Express)
```javascript
// auth.service.js
import bcrypt from 'bcryptjs';
import jwt from 'jsonwebtoken';
import { pool } from './db.js';
const registerUser = async (email, password) => {
// Validate input
if (!email || !password || password.length < 8) {
throw new Error('Invalid credentials');
}
// Check existing user
const existing = await pool.query('SELECT id FROM users WHERE email = $1', [email]);
if (existing.rows[0]) throw new Error('Email already registered');
// Hash password
const hashed = await bcrypt.hash(password, 12);
// Create user
const user = await pool.query(
'INSERT INTO users (email, password_hash) VALUES ($1, $2) RETURNING id, email',
[email, hashed]
);
// Generate verification token
const token = jwt.sign({ userId: user.rows[0].id }, process.env.JWT_SECRET, { expiresIn: '1h' });
return { userId: user.rows[0].id, email: user.rows[0].email, verificationToken: token };
};
export { registerUser };
```
### Validation (Jest + Supertest)
```javascript
// auth.service.test.js
describe('User Registration', () => {
beforeAll(() => {
// Setup test DB
});
it('should create user with valid credentials', async () => {
const result = await registerUser('[email protected]', 'validPassword123');
expect(result.userId).toBeDefined();
expect(result.verificationToken).toBeTruthy();
});
it('should reject short passwords', async () => {
await expect(registerUser('[email protected]', 'short')).rejects.toThrow('Invalid credentials');
});
it('should prevent duplicate emails', async () => {
await registerUser('[email protected]', 'password123');
await expect(registerUser('[email protected]', 'password456')).rejects.toThrow('Email already registered');
});
});
```
**Validation Results:**
- Code coverage: 100% (8/8 branches)
- All edge cases handled
- Follows ESLint Airbnb style guideApply to these tools
Browse all toolsSpeck
Automated meeting notes and action tracking
CoPilot AI
Hey, what’s on your mind today?
Microsoft Teams
Get more done every day with Microsoft Teams – powered by AI
GPT for work
Automate your spreadsheet tasks with AI power
Respell
Agentic AI Workflow platform
Notion
Connected workspace for docs, wikis, and projects
Compatible MCP servers
Browse all MCP servers
Find the right skills for your stack
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan