CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies

Rust client for Qdrant vector search engine

Semantic code search: tree-sitter parsing, local embeddings, and hybrid vector + BM25 retrieval. Multi-language, multi-backend.

Persistent memory for AI coding agents: Claude Code, Cursor, Windsurf, and Copilot. One command to install.

Compress code and text for LLMs by stripping noise, cutting tokens without changing the meaning.

MCP proxy to reduce LLM token costs, built in Rust

Local semantic search for Rust apps: store text, search by meaning, no cloud required.

Semantic cache proxy for MCP servers: reuse repeated responses to cut token usage.