Robot Resources Router
LLM routing proxy that cuts API costs 60-90% by auto-selecting the cheapest capable model across OpenAI, Anthropic, and Google. Provides stats, config, and model comparison tools via MCP.
README
@robot-resources/router-mcp
MCP server for managing Robot Resources Router — stats, config, and model comparison.
What is Robot Resources?
Human Resources, but for your AI agents.
Robot Resources cuts your LLM API costs by 60-90%. It runs a local proxy that classifies each prompt by task type (coding, reasoning, analysis, simple Q&A, creative, general) and routes it to the cheapest model that can handle it — across 14 models from OpenAI, Anthropic, and Google.
Your API keys never leave your machine. Free, unlimited, no tiers.
Install the full suite
npx robot-resources
One command sets up the Router proxy, configures your agents, and gets you saving immediately.
Learn more at robotresources.ai
About this MCP server
This package gives AI agents management tools for the Router proxy via the Model Context Protocol. Use it to check cost savings, compare models, and adjust routing config — all from within Claude Desktop, Claude Code, or any MCP-compatible agent.
Quick Start
npx @robot-resources/router-mcp
Requires the Router proxy running at localhost:3838 (default). Override with ROUTER_URL env var.
Configuration
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"router": {
"command": "npx",
"args": ["-y", "@robot-resources/router-mcp"]
}
}
}
Claude Code
Add to your .mcp.json:
{
"mcpServers": {
"router": {
"command": "npx",
"args": ["-y", "@robot-resources/router-mcp"]
}
}
}
Custom Router URL
If the proxy runs on a non-default port:
{
"mcpServers": {
"router": {
"command": "npx",
"args": ["-y", "@robot-resources/router-mcp"],
"env": {
"ROUTER_URL": "http://localhost:4000"
}
}
}
}
Tools
router_get_stats
Cost savings statistics for the Router proxy.
| Parameter | Type | Default | Description |
|---|---|---|---|
period |
"daily" | "weekly" | "monthly" |
"weekly" |
Time period |
task_type |
string |
— | Filter by task type |
provider |
string |
— | Filter by provider |
router_compare_models
Compare models by capability and cost for a given task type.
| Parameter | Type | Default | Description |
|---|---|---|---|
task_type |
string |
required | Task type to compare |
threshold |
number |
0.7 |
Minimum capability score |
provider |
string |
— | Filter by provider |
router_get_config
Read current routing configuration (provider scope, thresholds, overrides).
No parameters required.
router_set_config
Update routing configuration at runtime. Changes are in-memory only (reset on restart).
| Parameter | Type | Description |
|---|---|---|
provider_scope |
string |
Limit to specific provider(s) |
capability_threshold |
number |
Minimum capability score |
baseline_model |
string |
Model for cost comparisons |
log_level |
string |
Logging level |
Pass null for any field to reset it to the env-var default.
Requirements
- Node.js >= 18.0.0
- Router proxy running (see @robot-resources/router)
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.