pydantic-skills

Pydantic Skills

Logfire, Pydantic AI, and Pydantic plugins for Claude Code, Codex, and Cursor. Instrument your code with Logfire, query and export the resulting telemetry, build LLM agents with Pydantic AI, and validate data with Pydantic.

Plugins

Plugin	Hosts	Description	Capabilities
logfire	Claude Code, Codex, Cursor	Add Logfire observability and query/debug telemetry	Claude commands /instrument, /debug, /query; Codex/Cursor skills and MCP tools
logfire-exporter	Codex	Export Codex activity traces to Logfire	Codex lifecycle hooks
ai	Claude Code	Build AI agents with Pydantic AI	Pydantic AI skill docs
pydantic-ai-harness	Claude Code	Extend Pydantic AI agents with harness capabilities	Pydantic AI Harness skill docs
pydantic	Claude Code	Validate and serialize data with Pydantic	Pydantic skill docs

Install In Claude Code

Add this marketplace to Claude Code:

claude plugin marketplace add pydantic/skills

Then install a plugin:

claude plugin install logfire@pydantic-skills
claude plugin install ai@pydantic-skills
claude plugin install pydantic-ai-harness@pydantic-skills
claude plugin install pydantic@pydantic-skills

For Logfire, run /logfire:connect inside Claude Code to authenticate MCP and optionally configure native Claude telemetry. /logfire:status checks the two connections separately; /logfire:disconnect stops session capture without removing MCP or skills. See the connection and capture requirements.

Install In Codex

Add the published Pydantic marketplace to Codex:

codex plugin marketplace add pydantic/skills

Then open Codex's plugin UI and enable the plugins you want from the Pydantic marketplace:

• Logfire - installs Logfire skills and the hosted Logfire MCP server.
• Logfire Exporter - installs Codex lifecycle hooks that export completed Codex turns and tool calls to Logfire.

Configure Logfire Exporter with a Logfire write token in your environment or ${XDG_CONFIG_HOME:-~/.config}/logfire-exporter/config.env. Restart Codex after configuration, and run /hooks if Codex asks you to review or trust the new hooks.

To use the EU Logfire MCP endpoint in Codex without editing plugin files, replace the MCP entry and re-authenticate:

codex mcp remove logfire
codex mcp add logfire --url https://logfire-eu.pydantic.dev/mcp
codex mcp login logfire
codex mcp get logfire

Start a new Codex conversation after switching so the MCP tools reload.

Install In Cursor

Install the Logfire plugin from the published pydantic/skills repository:

git clone https://github.com/pydantic/skills.git
mkdir -p ~/.cursor/plugins/local
cp -R skills/plugins/logfire ~/.cursor/plugins/local/logfire

Then restart Cursor or run Developer: Reload Window. The plugin configures:

• display name Logfire
• Logfire skills for instrumentation, querying, and UI-opening workflows
• hosted Logfire MCP server from mcp.json

The Logfire MCP server requires normal Logfire authentication, such as logfire auth or a suitable LOGFIRE_TOKEN.

Cross-Agent Skills

The skills/ directory contains standalone SKILL.md files compatible with 30+ agents via the agentskills.io standard - including Codex, Cursor, Gemini CLI, and Claude Code.

Released skills are also published through the Agent Skills Discovery API at pydantic.dev/.well-known/agent-skills. After a change reaches main and passes this repository's sync check, CI asks pydantic/unified-docs to build and verify immutable, content-addressed artifacts from that exact commit. The generated discovery index and archives are deployment artifacts; do not maintain copies in this repository.

Skill	Description
logfire-setup	Set up Logfire by routing application, infrastructure, and eval projects
logfire-instrumentation	Add Logfire observability to Python, JS/TS, and Rust apps
logfire-infrastructure	Monitor hosts, containers, Kubernetes, databases, and cloud infrastructure
logfire-evals	Add dataset-based evaluations for AI and agent behavior
logfire-query	Query and analyze Logfire traces, logs, spans, metrics, and activity data
logfire-ui	Open Logfire project pages, live views, traces, and Explore filters
logfire-coding-agents	Record Claude/Codex sessions or configure coding-agent Gateway routing
logfire-frontend	Browser errors, Web Vitals, and session replays
logfire-gateway	Route application model calls through Pydantic AI Gateway
building-pydantic-ai-agents	Build LLM-powered agents with Pydantic AI — tools, capabilities, streaming, testing
pydantic-ai-harness	Extend Pydantic AI agents with harness capabilities: coding agents, Code Mode, sub-agents, memory
pydantic	Validate and serialize data with Pydantic models, constraints, and custom validators