Guardrail capabilities for Pydantic AI — cost tracking, prompt injection detection, PII filtering, secret redaction, tool permissions, and async guardrails. Built on pydantic-ai's native capabilities API.
Latest release 0.3.4 · 1 Jun 2026
2 files for Codex, OpenCode and Claude Code: pydantic-ai-shields AGENTS.md, pydantic-ai-shields CLAUDE.md — 799 tokens loaded in every session.
AGENTS.md A 417 tok CLAUDE.md A 382 tok These files are vstorm-co/pydantic-ai-shields's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.