agentops-mcp

Agentops MCP Server

agentops-mcp brings AgentOps trace and span data into MCP clients for developers debugging AI agent runs. Its tools authorize access, retrieve individual records and metrics, and fetch a complete trace when explicitly requested.

Server facts

agentops-mcp

Project

License
Not specified

Tech

Runtime / language
Not specified
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
14
Last commit
Data as of

Tags

  • Observability
  • Tracing
  • agents
  • ai

Overview

AgentOps run debugging centers on traces, spans, and the performance measurements attached to them. agentops-mcp exposes that observability data through MCP so developers can examine where an agent run succeeds or fails. The auth tool uses an AgentOps project API key for authorization; the documented authorization flow returns a JWT token. For targeted investigation, get_trace accepts a trace identifier, while get_span takes a span identifier, with both tools described as returning information and metrics.

When the investigation calls for the full record, get_complete_trace retrieves a trace together with its spans and associated metrics. Its tool description reserves that broader retrieval for explicit requests for complete or all data, distinguishing it from individual trace or span lookups. The server communicates over stdio. Setup documentation covers MCP client configuration, a Smithery installation route for Claude Desktop, and a local build workflow.

agentops-mcp setup notes

  • agentops-mcp records stdio only; confirm a network transport upstream before planning remote access.
  • agentops-mcp has no recorded license; confirm reuse terms in its repository.

Related MCP servers