argosvix/mcp-server

Overview

An MCP server that gives AI agents 89 tools for LLM observability. It queries cost, error, and latency data across 7 providers and can operate alerts, budgets, and evals from Claude or Cursor. Written in TypeScript.

Best for

Best for
Developers building AI agents who need cross-provider LLM cost and performance visibility.

Use cases

• Monitor LLM API costs and latency across providers
• Query error rates from Claude or Cursor
• Manage alerts, budgets, and evals through MCP tools

Notes

An MCP server that gives AI agents 89 tools for LLM observability. It queries cost, error, and latency data across 7 providers and can operate alerts, budgets, and evals from Claude or Cursor. Written in TypeScript.

0 stars on GitHub. Last updated 2026-08-30. Licensed MIT.

Use cases

• Monitor LLM API costs and latency across providers
• Query error rates from Claude or Cursor
• Manage alerts, budgets, and evals through MCP tools

Pros

• Large toolset with 89 observability functions
• Covers 7 LLM providers in one server
• Integrates directly with Claude and Cursor

Cons

• No GitHub stars yet, so adoption is unproven
• Focused only on observability, not broader agent tasks
• Requires an MCP-compatible client to use

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.