Foam Observability
Pull logs, traces, and metrics into Claude Code or Cursor while debugging production without leaving the agent.
Overview
Foam Observability is an MCP server for the Operate phase that lets your AI assistant query application logs, traces, and metrics through Foam's hosted API.
What is this MCP server?
- Streamable HTTP remote MCP at sdk.api.foam.ai with Bearer auth
- Query application logs, distributed traces, and metrics from one server
- Designed for AI coding assistants to investigate runtime behavior in context
- Foam Observability v1.0.0 per MCP server manifest
- No local stdio package—connect with a Foam account token only
- Server version 1.0.0 in manifest
- Single remote endpoint: sdk.api.foam.ai/mcp (streamable-http)
- Required secret header: Authorization Bearer token
What problem does it solve?
Debugging production issues forces constant tab-switching between your editor and observability UIs while the agent lacks real telemetry context.
Who is it for?
Indie builders and tiny teams already on Foam who want MCP-native observability inside Claude Code, Cursor, or similar agents.
Skip if: Teams without Foam instrumentation or builders who need on-prem-only, self-hosted log stacks with no cloud token.
What do I get? / Deliverables
Your agent can pull Foam-backed logs, traces, and metrics in-thread so you diagnose incidents faster without leaving the coding workflow.
- Agent-queryable access to Foam logs, traces, and metrics
- Faster in-editor production diagnosis loops
Recommended MCP Servers
Journey fit
Observability data is consumed after ship when apps run in production and you need to diagnose incidents or regressions. Foam exposes unified telemetry queries—classic monitoring work rather than feature build or launch distribution.
How it compares
Hosted observability MCP integration, not a local lint skill or generic web scraper.
Common Questions / FAQ
Who is Foam Observability for?
Solo and small-team developers who use Foam for telemetry and want their AI coding agent to read logs, traces, and metrics during incident response.
When should I use Foam Observability?
Use it in Operate when production misbehaves, after deploys, or when you need correlated traces and metrics while fixing bugs with an agent.
How do I add Foam Observability to my agent?
Register the remote MCP URL https://sdk.api.foam.ai/mcp in your client and set Authorization to a Bearer token from your Foam account.