Monitor AI agent spending, detect cost anomalies, and optimize model selection across 10 domains with 23 built-in tools. For teams running multi-agent systems who need visibility into LLM costs and ROI.
io.github.metrxbots/mcp-server
Local install
STDIO
1 required env var
How models use it and what it is built for.
Monitor AI agent spending, detect cost anomalies, and optimize model selection across 10 domains with 23 built-in tools. For teams running multi-agent systems who need visibility into LLM costs and ROI.
Local install — runs as a subprocess.
Configuration this server reads at startup.
Your Metrx API key (sign up free at https://app.metrxbot.com/sign-up — manage keys at https://app.metrxbot.com/settings/security)
Where to find authoritative docs and source for Metrxbots.
Paste any of these into Agent Studio after connecting Metrxbots.
Common questions about connecting and running Metrxbots.
What do I need to get started with this MCP server?
Sign up free at https://app.metrxbot.com/sign-up to get a METRX_API_KEY, then set that environment variable when running the server. The server connects via stdio and exposes 23 tools across cost tracking, anomaly detection, and optimization domains.
Can this server track costs for multiple AI models?
Yes. The server is designed to monitor spending across different models and agents, allowing you to compare cost efficiency and optimize model selection based on your actual usage patterns.
Does this MCP server require a paid Metrx plan?
You can sign up free at https://app.metrxbot.com/sign-up. Pricing details and plan tiers are available on the Metrx website; check your account dashboard to manage API keys and billing.
What happens if my METRX_API_KEY is invalid or missing?
The METRX_API_KEY environment variable is required to authenticate with the Metrx backend. If it's missing or invalid, the server will fail to initialize. Verify your key in the Metrx dashboard and ensure it's set before starting the MCP server.
How does this compare to manual cost tracking or other monitoring tools?
This MCP server integrates cost tracking, anomaly detection, and ROI analysis into your AI workflow via 23 specialized tools, eliminating manual log parsing. It's built specifically for multi-agent systems and provides automated insights rather than static dashboards.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent StudioJudges Panel
45 judges that evaluate AI-generated code for security, cost, and quality with built-in AST.
Addozhang Nexus
Query Sonatype Nexus Repository (OSS/Pro) for Maven, Python, and Docker artifacts
Cyanheads MCP Ts Template
TypeScript template for building MCP servers with declarative tooling, observability, and auth.
Chernistry Bernstein
Declarative agent orchestration for engineering teams