Discover and explore top open-source AI tools and projects—updated daily.
NSHipsterAI-readable Apple documentation service
Top 91.6% on SourcePulse
Summary
sosumi.ai addresses the challenge of making Apple's extensive developer documentation accessible to AI and machine processing. It provides a service that converts JavaScript-rendered Apple Developer documentation pages, including API references and Human Interface Guidelines, into a machine-readable Markdown format. This benefits engineers, researchers, and power users by enabling programmatic access and analysis of crucial Apple platform information.
How It Works
The core approach involves a web service that intercepts requests for Apple Developer documentation URLs. By replacing developer.apple.com with sosumi.ai, users can access Markdown versions of the content. The service fetches the original JavaScript-rendered pages, processes them into Markdown, and serves the result. It is built using Hono, making it compatible with various runtimes like Cloudflare Workers, and includes an MCP server supporting Streamable HTTP and SSE for direct client integration or stdio proxying.
Quick Start & Requirements
git clone https://github.com/nshipster/sosumi.ai.git), navigate into the directory (cd sosumi.ai), and install dependencies (npm install).npm run dev. The service will be available at http://localhost:8787 by default.Highlighted Details
developer.apple.com with sosumi.ai in any Apple Developer documentation URL.doc:// URI Scheme: Provides a custom scheme (doc://{path}) for direct access to Markdown content.searchAppleDocumentation (structured results) and fetchAppleDocumentation (Markdown content).Maintenance & Community
The README does not specify notable contributors, sponsorships, or community channels such as Discord or Slack.
Licensing & Compatibility
Limitations & Caveats
This is an unofficial project, not affiliated with or endorsed by Apple Inc. Users are solely responsible for ensuring their use complies with Apple's Terms of Use and relevant laws. The service provides transient content, fetching and briefly caching it on demand without permanent archiving, and implements rate limiting to avoid excessive load. It does not bypass security measures or facilitate redistribution.
2 days ago
Inactive