Getting Started
AI Coding Stack is a comprehensive directory and metadata repository for the AI coding ecosystem. This guide will help you understand how to use the website and contribute to the project.
Exploring the Stack
Browse Categories
Visit aicodingstack.io to explore different categories of AI coding tools:
- IDEs - Find AI-powered integrated development environments (VS Code, Cursor, TRAE)
- CLIs - Explore command-line AI coding assistants (Codex, Claude Code)
- Models - Compare large language models for coding (Kimi K2, DeepSeek V3.1, GLM 4.5, Qwen3 Coder)
- Providers - Review LLM API providers (DeepSeek, Moonshot, SiliconFlow, OpenRouter)
Compare Tools
Use the website to compare different tools side-by-side, reviewing their features, capabilities, and official documentation links.
Contributing to the Metadata
AI Coding Stack is community-maintained. You can contribute by adding new tools or updating existing entries in our manifest files.
Understanding Manifests
All tool metadata is stored in JSON files under the manifests/ directory:
ides.json- IDE metadataclis.json- CLI tool metadatamodels.json- LLM model metadataproviders.json- API provider metadata
Adding a New Tool
- Clone the repository
- Locate the appropriate manifest file in
manifests/ - Add your entry following the existing schema
- Ensure all URLs are accessible
- Submit a pull request
Manifest Schema Examples
For IDEs/CLIs:
{
"name": "Tool Name",
"id": "tool-id",
"vendor": "Vendor Name",
"description": "Brief description",
"websiteUrl": "https://...",
"docsUrl": "https://...",
"latestVersion": "1.0.0"
}
For Models:
{
"name": "Model Name",
"vendor": "Vendor",
"id": "model-id",
"size": "70B",
"contextWindow": 128000,
"maxOutput": 4096,
"pricing": {...},
"urls": {...}
}
Next Steps
- Browse the AI Coding Stack to discover tools
- Check the
manifests/directory to see how metadata is structured - Read the CLAUDE.md for detailed contribution guidelines
- Submit your first contribution to help grow the directory