logfire Claude Code Plugin
Claude Code PluginClaude CodeMonitoringMonitoring & ObservabilityWeb & App DevelopmentAI Agents & AI App DevelopmentThe explanation below is AI-generated. Please verify it against the sources.
This is a Claude Code (also Codex and Cursor) plugin that adds Pydantic Logfire observability integration for Python codebases. According to its README, it includes a command that scans a project and auto-adds instrumentation for frameworks such as FastAPI, httpx, asyncpg, and SQLAlchemy, plus several skills for setup routing, infrastructure monitoring, and evaluation-dataset creation. It also provides commands to investigate errors and query telemetry through an MCP server connection to the Logfire service, and a skill to open Logfire's hosted UI pages directly. The plugin bundles a reference document describing core Logfire configuration and instrumentation patterns. It supports installation across Claude Code, Codex, and Cursor, and can target either the US or EU Logfire region or a self-hosted instance.
About the service
Logfire is an observability service operated by Pydantic. Per its homepage, it ingests traces, logs, and metrics over OpenTelemetry from Python, JavaScript/TypeScript, Rust, and other OTel-instrumented sources, letting a team inspect model calls, API requests, and database queries together in a single trace. It offers a free tier of 10 million records per month, paid tiers with additional usage billed per record, and an enterprise tier with self-hosted deployment. The homepage also states it holds SOC 2 Type II certification, supports HIPAA business associate agreements, and offers an EU data region, alongside a Postgres-style query interface and evaluation features for AI/agent traces.
What you can do with logfire
- Scan a Python project and auto-add Logfire instrumentation for supported frameworks with a dedicated command
- Run a setup skill that authenticates and routes to the correct configuration workflow for the repository
- Investigate errors by pulling Logfire traces via MCP with a debug command
- Query traces, logs, and metrics interactively or embed query logic in code with a query command
- Reference a bundled document covering core Logfire configuration, instrumentation, structured logging, and AI/LLM tracing patterns