munhq

Popular repositories Loading

1. codeindex codeindex Public

   Structural code-intelligence engine in Zig — tree-sitter across 40+ languages, trigram and inverted indexes, dependency graph, exposed as an MCP server

   Zig 3

2. chat-recall chat-recall Public

   One memory across every AI coding tool. Indexes Claude Code, Gemini CLI, Codex, OpenCode and Antigravity sessions into a single searchable history, and exposes it to your agent over MCP.

   TypeScript 2

3. distil distil Public

   Measure what context compression actually costs an LLM agent: where a session's tokens go, and whether a rewrite pays for the prompt cache it invalidates. Rust crate, MCP server and benchmark harness.

   Rust 1

4. gpuscale gpuscale Public

   Run LLM inference on the cheapest GPU anywhere — a Kubernetes controller that buys spot GPUs across 7 providers, serves vLLM on them, and scales to zero

   Go

5. cloud-tools cloud-tools Public

   Multi-cloud cost, inventory and waste analysis in Rust — AWS, GCP, Cloudflare and OVH, as an MCP server and an HTTP API

   Rust

6. gpu-api-ray gpu-api-ray Public

   OpenAI-shaped GPU batch-inference API in Go — dashboard, Dragonfly-backed job store, Ray Serve or direct vLLM workers, Prometheus metrics

   Go