ConfigsApr 1, 2026·2 min read

OpenSpec — Spec-Driven AI Development

OpenSpec provides structured specifications that AI coding agents follow to produce consistent code. 36K+ stars. Works with Cursor, Claude Code, Copilot. MIT.

TO
TokRepo精选 · Community
Quick Use

Use it first, then decide how deep to go

This block should tell both the user and the agent what to copy, install, and apply first.

1. Install the VS Code extension or use the CLI: ```bash npm install -g openspec openspec init ``` 2. This creates an `openspec.yaml` in your project root: ```yaml project: name: "my-app" stack: ["Next.js", "TypeScript", "Tailwind"] specs: - name: "User Authentication" description: "OAuth2 login with Google and GitHub" requirements: - "Use NextAuth.js v5" - "Support Google and GitHub providers" - "Store sessions in database" constraints: - "No client-side token storage" - "CSRF protection enabled" ``` 3. Point your AI agent at the spec: ``` Implement the "User Authentication" spec from openspec.yaml ``` The AI agent reads the structured spec and produces code that matches all requirements and constraints.
Intro
OpenSpec is a specification-driven development (SDD) framework for AI coding assistants. Instead of writing long, ambiguous natural language prompts, you define structured specs that any AI tool can interpret consistently. Key benefits: - **Consistent output**: Same spec produces similar code regardless of which AI tool you use - **Reviewable**: Specs are version-controlled, reviewable, and diffable — unlike chat history - **Composable**: Break large features into small specs, implement incrementally - **Tool-agnostic**: Works with Cursor, Claude Code, GitHub Copilot, Codex, Windsurf - **Testable**: Specs include acceptance criteria that can be verified automatically Think of it as: `.cursorrules` tells the AI HOW to code, OpenSpec tells it WHAT to build.
## FAQ **Q: How is this different from .cursorrules or CLAUDE.md?** A: Those define coding style and conventions. OpenSpec defines features and requirements. They're complementary — use both. **Q: Does the AI agent need special support for OpenSpec?** A: No. Any AI that can read YAML/Markdown can follow OpenSpec files. The structure is self-explanatory. **Q: Can I use it for existing projects?** A: Yes. Run `openspec init` and add specs for new features. No need to spec existing code. **Q: How detailed should specs be?** A: Enough that two different developers (or AI agents) would build roughly the same thing. Include requirements, constraints, and acceptance criteria. ## Works With - Claude Code, Cursor, GitHub Copilot, Codex, Windsurf - Any AI coding agent that reads project files - VS Code (official extension) - Git (specs are version-controlled YAML files)
🙏

Source & Thanks

- GitHub: https://github.com/Fission-AI/OpenSpec (36K+ stars) - License: MIT - Docs: https://openspec.dev

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.

Related Assets

LaVague — Natural Language Web Automation

Give a text objective, LaVague drives the browser to accomplish it. Large Action Model framework for web agents. 6.3K+ stars.

TokRepo精选

Trae Agent — AI Coding Agent by ByteDance

Open-source autonomous coding agent for software engineering tasks. Multi-provider LLM support. By ByteDance. 11K+ stars.

TokRepo精选

bolt.diy — AI Full-Stack App Builder, Any LLM

Community fork of Bolt.new. Prompt, edit, and deploy full-stack web apps with any LLM provider. 19K+ GitHub stars.

TokRepo精选
◈Home🔍Search👤Me