mcp-search-proxy hides 532 MCP tools (~179k tokens/turn) behind 4 tools via BM25
pisa_p · reddit · 2026-09-12
The author aggregates 21 MCP servers (532 tools) behind LiteLLM for OpenWebUI, but full schemas injected every turn cost 179k tokens even when no tool is needed.
Fix: mcp-search-proxy sits in between and exposes just 4 tools:
- search: BM25 over a cached tool manifest
- describe: fetch full schemas for up to 10 tools
- call: route to the right upstream
- refresh: rebuild the index
Results: same capability at 60x less context (2-3k tokens); 560 tools fit in 1.8k chars. Pure local BM25 — no embeddings, no LLM inside. Per-upstream sticky sessions keep a stateful browser server working across navigate/scan/snapshot.
When to use: worth it with 100+ tools where most turns need 0-1; not under 30 tools; lexical only, no semantic search. Docker Compose, MIT, open-sourced on GitHub.
More from coding & agent
- Anthropic team uses Claude for on-call: alert triage, root cause and fix proposals — ClaudeDevs · 2026-09-12
- Developer streams an AI agent playing Minecraft all day via Codex, aiming to build an auto chicken farm — nickbaumann_ · 2026-09-12
- Why working with AI agents all day is exhausting: delegation adds mental load, not removes it — bendee983 · 2026-09-12
- "Codex is my main interface to the system": user says agent made Linux admin effortless — mark_k · 2026-09-12
- Alchemy's distilled project builds unified agent-friendly SDKs for 60+ SaaS services — samgoodwin89 · 2026-09-12
- RL on just 1,700 tasks lifts Kimi K2.7 across five coding benchmarks — echen · 2026-09-12