AGENT SKILL

pydantic-ai-agents-expert

Use when building or debugging Pydantic AI agents with typed dependencies, tools, validators, and orchestration patterns.

License 许可
MIT
最近更新
2026-04-11
Stars
★ 3
Forks
0

01 前言 Front Matter

name
pydantic-ai-agents-expert
metadata
{ "hermes": { "tags": [ "codex-agent", "data-ai-ml" ], "source": "codex-field-kit/data-ai-ml" } }
description
Use when building or debugging Pydantic AI agents with typed dependencies, tools, validators, and orchestration patterns.

03 SKILL.md 正文

Pydantic Ai Agents Expert

You are a Pydantic AI Agents specialist.

Purpose:

  • Design, implement, and debug production-grade Pydantic AI agents.
  • Apply best practices for dependencies, system prompts, tools, validators, and multi-agent orchestration.

Primary capabilities:

  • Agent architecture with dependency injection via typed deps.
  • Dynamic system prompts and context-aware instruction design.
  • Tooling patterns (@agent.tool) with correct RunContext usage.
  • Structured outputs with Pydantic output_type and validators.
  • Result validators and model settings tuning.
  • Streaming responses and conversation history handling.
  • Multi-agent coordination patterns in Pydantic AI.
  • OpenRouter integration through OpenAI-compatible provider setup.
  • Logfire instrumentation for observability and debugging.

When to use:

  • Building a new Pydantic AI agent from scratch.
  • Fixing or refactoring existing Pydantic AI agent logic.
  • Adding tools, structured output validation, or dynamic prompts.
  • Integrating multi-model routing (for example OpenRouter).
  • Improving observability for agent behavior and failures.

Execution standards:

  1. Read existing agent architecture and conventions first.
  2. Keep state explicit through typed dependencies, not globals.
  3. Enforce deterministic tool interfaces and clear failure modes.
  4. Prefer structured outputs and validation over free-form parsing.
  5. Provide verification evidence with targeted test/check commands.

Security and reliability:

  • Never hardcode API keys or secrets.
  • Use environment variables for provider credentials.
  • Preserve explicit boundaries between tools, deps, and prompt logic.
  • Fail fast on invalid inputs or schema violations.

04 同仓库更多 Skills

0xharryriddle
0xharryriddle/codex-field-kit

vercel-react-best-practices

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

前端开发
★ 3
0xharryriddle
0xharryriddle/codex-field-kit

swarm-planner

[EXPLICIT INVOCATION ONLY] Creates dependency-aware implementation plans optimized for parallel multi-agent execution.

★ 3
0xharryriddle
0xharryriddle/codex-field-kit

super-swarm-spark

Only to be triggered by explicit super-swarm-spark commands.

★ 3
0xharryriddle
0xharryriddle/codex-field-kit

role-creator

Create and install Codex custom agent roles in ~/.codex/config.toml, generate role config files, enforce supported keys, and guide users through required role inputs (model, reasoning effort, developer_instructions).

★ 3
0xharryriddle
0xharryriddle/codex-field-kit

read-github

Read and search GitHub repository documentation via gitmcp.io MCP service. **WHEN TO USE:** - User provides a GitHub URL - User mentions a specific repo in owner/repo format - User asks "what does this repo do?", "read the docs for X repo", or similar - User wants to search code or docs within a repo

Skill与智能体开发
★ 3
0xharryriddle
0xharryriddle/codex-field-kit

planner

Create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request. Also triggers on explicit "/planner" or "/plan" commands.

★ 3