mcp-scheduling
A framework-agnostic tool router for MCP (Model Context Protocol) servers. Routes agent requests to the fastest available tool when similar tools exist, decreasing response latency by ~64% (tool dependent) on average across 824 benchmark questions.
Tool similarity is determined by computing vector embeddings of each tool's description, then grouping duplicates via cosine similarity, letting the router identify functionally equivalent tools automatically.
Tested against Google's Frames dataset using Gemini Flash with a custom agent built using LangChain.
README
mcp-scheduling
it's actually a router
A framework-agnostic tool router for MCP (Model Context Protocol) servers. Routes agent requests to the fastest available tool when similar tools exist, cutting response latency by ~64% (tool dependent) on average across 824 benchmark questions.
Tool similarity is determined by computing vector embeddings of each tool's description, then grouping duplicates via cosine similarity. This lets the router identify functionally equivalent tools automatically.
The router is framework agnostic, not dependent on ReAct.
Testing Methodology
NOTE
Names of the MCP servers/tools are generic when provided to the agent. No mention of slow or fast.
Agent uses google/gemini-2.0-flash-001 and is provided two MCP servers:
scrape-slow (5 second delay)
scrape-fast (no delay)
The agent response verifier used openai/gpt-oss-120b.
The scheduler was tested against all 824 questions in the Google Frames Dataset.