MCP Cost Optimizer runs on your own machine — the client starts it, so there is no endpoint to ping. 533 installs a week from npm. Last commit 9 Oct 2026.
Local-first LLM cost analysis and model-switch savings estimates for AI agents.
Today is the operative word: we check MCP Cost Optimizer every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add mcp-cost-optimizer -- npx -y mcp-cost-optimizer
{
"mcpServers": {
"mcp-cost-optimizer": {
"args": [
"-y",
"mcp-cost-optimizer"
],
"command": "npx"
}
}
}
[mcp_servers.mcp-cost-optimizer]
command = "npx"
args = ["-y", "mcp-cost-optimizer"]
{
"mcpServers": {
"mcp-cost-optimizer": {
"args": [
"-y",
"mcp-cost-optimizer"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"mcp-cost-optimizer": {
"args": [
"-y",
"mcp-cost-optimizer"
],
"command": "npx"
}
}
}
Local-first personal memory and model server for trusted MCP agents on macOS.
Local-first persistent memory MCP server for AI coding agents.
Local-first codebase intelligence MCP server for AI coding agents.
Real eyes and hands on the web for AI agents. MCP-first. Local-first. Vendor-agnostic.
Local-first document management and semantic search for AI coding agents
Local BYOK MCP for AI model shortlisting using OpenRouter + Artificial Analysis.
Local-first web intelligence MCP server for AI coding agents
Analyze LLM API costs: token waste detection, caching savings estimates, model comparison
Answers built from our own checks of this server.