Spec-Kitty automates software development workflows using AI agents. It enables developers to manage coding tasks through spec-driven development, Kanban dashboards, and git worktrees. Integrates with Claude, Cursor, Gemini, and Codex for efficient feature shipping.
git clone https://github.com/Priivacy-ai/spec-kitty.gitSpec-Kitty is a development automation tool designed for teams using AI agents like Claude, Cursor, Gemini, and Codex. It enables developers to manage coding tasks through spec-driven development methodologies, combining structured specifications with AI-powered code generation. The tool provides Kanban dashboards for task visualization and git worktree management for organized branch handling. By automating software development workflows, Spec-Kitty helps teams ship features more efficiently while maintaining code organization and clarity around task requirements.
1. **Initialize the Project**: Run `spec-kitty init [PROJECT_NAME] --stack [TECH_STACK]` in your terminal to create a new project directory with basic structure. 2. **Define the Feature**: Use the prompt template above, replacing placeholders with your specific requirements (e.g., `[FEATURE_NAME]`, `[TECH_STACK]`). 3. **Generate Assets**: Execute the generated commands to create the git worktree and Kanban board. Copy the specification document into your project's `docs/` folder. 4. **Integrate with IDE**: In your preferred IDE (Cursor/CLion), use the spec document to guide implementation. Enable the AI assistant's context-aware mode to reference the spec during coding sessions. 5. **Track Progress**: Update the Kanban board daily with task completion status. Use `spec-kitty status` to generate a progress report for stakeholders. **Tips:** - For complex features, break them into smaller sub-tasks in the Kanban board. - Use `spec-kitty review` to validate the specification against best practices before implementation. - Integrate with GitHub Actions using the generated workflow templates for automated testing.
Automate the generation of executable specifications for new features.
Track project progress in real-time using a live Kanban dashboard.
Orchestrate multiple AI agents to work on different parts of a project simultaneously.
Prevent merge conflicts through automated validation and pre-flight checks.
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/Priivacy-ai/spec-kittyCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Act as Spec-Kitty to automate the development of [FEATURE_NAME] for [PROJECT_NAME]. Create a detailed technical specification document covering [TECH_STACK] requirements, API endpoints, and database schema. Generate a Kanban board with [NUMBER_OF_TASKS] tasks in GitHub/GitLab. Set up a git worktree for [BRANCH_NAME] and provide the first commit message with the specification. Include error handling and edge cases in the spec. Use [DEVELOPMENT_METHODOLOGY] for task prioritization.
### Technical Specification: Real-time Chat Notifications for Acme CRM **Overview:** Implement a WebSocket-based notification system for CRM users when new messages arrive in their inbox. The system will push real-time updates to the frontend without requiring page refreshes. **Tech Stack:** Node.js (v20), Express.js, Socket.IO, PostgreSQL (v15), Redis (for pub/sub), React (frontend) **API Endpoints:** 1. `POST /api/notifications/subscribe` - Subscribes a user to their notification channel 2. `GET /api/notifications` - Returns paginated list of unread notifications 3. `PUT /api/notifications/read` - Marks notifications as read **Database Schema:** ```sql CREATE TABLE notifications ( id SERIAL PRIMARY KEY, user_id INTEGER REFERENCES users(id) ON DELETE CASCADE, message TEXT NOT NULL, is_read BOOLEAN DEFAULT FALSE, created_at TIMESTAMP DEFAULT NOW(), channel VARCHAR(50) NOT NULL ); ``` **Kanban Board (GitHub Projects):** - Backend Setup (High Priority) - WebSocket Integration (High Priority) - Frontend Notification UI (Medium Priority) - Database Migration Scripts (Low Priority) - Error Handling & Retry Logic (Medium Priority) - Load Testing (Low Priority) **Git Worktree Setup:** ```bash git worktree add -b feature/chat-notifications feature/chat-notifications cd feature/chat-notifications ``` **First Commit Message:** "feat: initial spec for real-time chat notifications - Defines WebSocket architecture - Specifies API contracts - Includes error handling scenarios - Links to Figma mockups for UI" **Edge Cases to Handle:** 1. User disconnects mid-conversation (auto-reconnect with exponential backoff) 2. High volume of simultaneous notifications (rate limiting per user) 3. Database connection failures (queue notifications in Redis for later delivery) 4. Frontend offline scenarios (sync notifications when connection restored) **Development Methodology:** Scrum with 2-week sprints, daily standups at 10 AM EST.
AI-powered code editor by Codeium
Create and collaborate on interactive animations with powerful, user-friendly tools.
Google's multimodal AI model and assistant
The AI Code Editor for productive developers
AI assistant built for thoughtful, nuanced conversation
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan