mcpspend-cost-tracking

Track MCP tool call costs per tool, project, and customer via a proxy wrapper.

7|2|Updated May 16, 2026
One-click install
npx skills add https://github.com/reason-machines/mcp-skills --skill mcpspend-cost-tracking-reason-machines
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mcpspend-cost-tracking
Source: https://github.com/reason-machines/mcp-skills/tree/main/skills/mcpspend-cost-tracking
Command: npx skills add https://github.com/reason-machines/mcp-skills --skill mcpspend-cost-tracking-reason-machines

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @mcpspend/proxy, @mcpspend/mcp-server.

What problem does it solve? Teams running MCP servers across Claude Desktop, Cursor, Windsurf, and VS Code have no visibility into how much each tool call costs, making it impossible to attribute AI agent spending to specific tools, projects, or customers. ## Core Features & Use Cases - Transparent Proxy Wrapping: Automatically wraps any MCP server (stdio or HTTP/SSE) to record per-call cost, latency, and token metadata without changing server behavior. - Usage Query MCP Server: Exposes tools like get_today_cost, list_top_tools, get_tool_usage, and list_sessions so you can query spending directly from any MCP client. - Per-Customer Attribution & Budget Alerts: Attach metadata such as customer and project to each server, then configure thresholds and Slack/email alerts in the dashboard. - Use Case: An agency running MCP servers for multiple clients wraps each server with customer metadata, then queries monthly spend per customer to generate accurate invoices. ## Quick Start Ask the AI to set up MCPSpend cost tracking by running the init command with your API key to wrap all configured MCP servers.

Frequently Asked Questions about mcpspend-cost-tracking

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I track MCP tool call costs in Claude Desktop?▼

Run npx @mcpspend/proxy init with your API key to auto-detect Claude Desktop and wrap all configured MCP servers. The proxy records cost, latency, and token metadata per call and reports it to your MCPSpend dashboard.

How to query MCP usage data from within an MCP client?▼

Add the @mcpspend/mcp-server to your client config with your API key. It exposes tools like get_today_cost, list_top_tools, get_tool_usage, and list_sessions for querying spend directly in chat.

Does MCPSpend work with Cursor, Windsurf, and VS Code?▼

Yes, the init command auto-detects Claude Desktop, Cursor, Windsurf, and VS Code configs and wraps their servers. You can also specify a client explicitly with the --client flag or pass a custom config path.

Can MCPSpend wrap remote HTTP or SSE MCP servers?▼

Yes, use the wrap-http command with the remote server URL and your API key. Alternatively, query usage remotely through the hosted HTTP MCP endpoint at api.mcpspend.com using Bearer token authentication.

Why is the MCPSpend proxy not reporting usage data?▼

Verify your API key against the health endpoint with curl, confirm the client config contains mcpspend entries, and check that the .mcpspend.bak backup exists. Set MCPSPEND_DEBUG=1 and restart the client for detailed logs.

What are the MCPSpend free tier limits?▼

The free tier allows 25K tracked calls per month. You can check current usage with the usage command, set budget alerts in the dashboard, or upgrade on the pricing page when the limit is exceeded.