PydanticAI. Type-Safe Verified Memory.

Type validation meets fact verification.

Get API Key →

12MCP tools

0.01mslookup

Freefor individuals

ed25519signed

The Problem

PydanticAI validates the shape of data. But it doesn't validate truth. A well-structured response can still be wrong. Type-safe doesn't mean truth-safe.

Step 1

Connect

→

Step 2

Ingest + Extract

→

Step 3

Verify + Receipt

The Fix

npm install @verifiedstate/pydantic-ai

Beta: tagging-level integration. Ingestion works via @verifiedstate/sdk with PydanticAI-specific source tagging. Native PydanticAI features coming.

textBeta
// Use the REST API directly from PydanticAI agents
// POST /ingest → POST /extract → POST /verify

What You Get Back

✓ VERIFIED

receipt.json
{
  "receipt_id": "rct_7f3a...",
  "assertion": "User prefers dark mode",
  "status": "verified",
  "confidence": 0.94,
  "signature": "ed25519:a8c3f1...",
  "verified_at": "2026-04-06T14:32:01Z",
  "chain_position": 42
}

Without VerifiedState vs. With VerifiedState

Feature	Without	With VerifiedState
Data validation	Shape only	Shape + truth verification
Memory	No persistence	Verified persistent store

FAQ

Is there a Python SDK?

Not yet. Use the REST API. Python SDK is on the roadmap.

Related Integrations

Start building with verified memory.

Free: 50,000 assertions/month. No credit card. No trial expiration.

Get Free API Key Read the Docs GitHub →