MCPSpend
Track MCP tool call costs, latency, and errors across AI agent sessions with budget alerts and attribution.
MCPSpend is a transparent proxy that wraps MCP servers and records every tool call — tracking latency, errors, payload size, and per-model cost estimation. Teams use it to monitor AI spending, attribute costs per team or customer, set budget alerts, and export usage data to analytics warehouses like BigQuery and Snowflake.
Source
Repository: https://github.com/andreisirbu91-lab/mcpspend
Maintain MCPSpend?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[MCPSpend on getagentictools](https://getagentictools.com/mcp/andreisirbu91-lab-mcpspend?ref=badge) Related mcp
Misata Generates realistic, multi-table synthetic datasets from natural language descriptions, YAML schemas, or existing databases with r
OneCite Generate academic citations from DOI, arXiv, titles, or URLs in BibTeX, APA, MLA formats
Ollama PostgreSQL Data Analysis Integrates natural language querying with PostgreSQL databases to enable AI-interpreted data analysis without manual SQL writing.
Temporal Workflows Exposes Temporal workflows as tools for AI assistants, enabling organizations to leverage existing workflow infrastructure without
Sling Data integration tool for querying databases, running data pipelines, and managing replications across 30+ systems
Appwrite Backend-as-a-service platform integration for database, user, storage, and function management