The Agent Skills Directory

[SAFE]: The skill serves as a developer migration guide and does not contain any malicious code, prompt injections, or unauthorized data access patterns. It provides clear technical instructions and refers only to official, well-known documentation and repositories.
[SAFE]: Python code examples provided in the references are designed for testing and demonstration purposes, utilizing standard libraries such as asyncio and tempfile, along with vendor-provided packages pydantic-ai and pydantic-ai-harness.
[SAFE]: The skill includes explicit security recommendations, such as enforcing authorization and isolation at the backing service level and treating prompt instructions and path checks as defense-in-depth measures.

migrating-deep-agents-to-pydantic-ai-harness