# mo_creates

- **Event:** [Built with Opus 4.7: a Claude Code hackathon](https://cerebralvalley.ai/e/built-with-4-7-hackathon)
- **When:** Apr 21 at 12:00 PM – Apr 27 at 2:00 AM (EDT)
- **Where:** Online
- **Team:** [Muhammed Nehan](https://cerebralvalley.ai/u/muhnehh)
- **GitHub:** https://github.com/muhnehh/loom-mcp
- **Demo video:** https://www.youtube.com/watch?v=z1gBFO4jdbI
- **Gallery:** https://cerebralvalley.ai/e/built-with-4-7-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/built-with-4-7-hackathon/hackathon/gallery/121

LoomMCP — The context compiler for AI coding agents.

Every time Claude Code opens your codebase, it reads thousands of lines it doesn't need. A typical session burns 140,000+ tokens just on code reading - most of it irrelevant context. That's money wasted, latency added, and quality degraded.

LoomMCP fixes this at the infrastructure layer.

It's a single MCP server that indexes your codebase once using tree-sitter AST parsing, then lets Claude retrieve only the exact code it needs -- function signatures, specific implementations, references, dependencies, with byte-level precision.

---

Markdown version of https://cerebralvalley.ai/e/built-with-4-7-hackathon/hackathon/gallery/121. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt