Metadata-Version: 2.4
Name: pydanticai-zambo
Version: 0.2.5
Summary: Verifiable execution receipts for PydanticAI agents. Every tool call mints a verifiable AER-1 receipt at zambo.dev.
Author: Zambo
License: MIT
Project-URL: Homepage, https://zambo.dev
Keywords: pydanticai,ai agent receipt,verifiable receipt,observability,mcp,verifiable-execution-receipts,execution-receipts,verifiable-receipt
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# pydanticai-zambo

Verifiable AER-1 execution receipts for PydanticAI agents. PydanticAI function tools for [Zambo](https://zambo.dev), with automatic verifiable receipt capture.

Receipts follow [AER-1](https://zambo.dev/aer-1/), the verifiable execution receipt standard created by Brennan Zambo. Every `ask_zambo` call routes a goal through Zambo's tools over the public MCP endpoint, and every call mints a public verifiable receipt at `https://zambo.dev/run/{id}`.

## Install

```bash
pip install pydanticai-zambo
```

## Ten-line example

```python
from pydanticai_zambo import ZamboToolset
from pydantic_ai import Agent
from pydantic_ai.models.test import TestModel

toolset = ZamboToolset()
agent = Agent(TestModel(), tools=toolset.tools())
result = agent.run_sync("Summarize the Zambo free tier")
toolset.record_run(result.output)
print(toolset.last_receipt_url)
```

The tool prints each public `https://zambo.dev/run/{id}` URL and keeps the most recent URL in `last_receipt_url`. The receipt describes what the endpoint recorded and observed. It does not prove an outside outcome that the endpoint did not observe.

Set `ZAMBO_API_URL` to point at a compatible Zambo host, or `ZAMBO_TIMEOUT` to change the request timeout. Receipt failures land in `receipt_errors` and never break the host agent.

The package imports cleanly without `pydantic-ai` installed; calling `tools()` without it raises an informative `ImportError` telling you to `pip install pydantic-ai`.

Project homepage: [https://zambo.dev](https://zambo.dev)

AER-1 receipt spec: [https://zambo.dev/aer-1/](https://zambo.dev/aer-1/)

## Guides

- [Execution Receipts for AI Agent Frameworks](https://zambo.dev/guides/framework-agent-execution-receipts/)
- [What Is an AI Agent Execution Receipt?](https://zambo.dev/guides/ai-agent-execution-receipt/)

## Local AER-1 receipts (no network needed)

The package also mints AER-1 execution receipts locally, with no API calls:

```python
from pydanticai_zambo import mint_receipt, verify_receipt

receipt = mint_receipt("my_tool", {"query": "hello"}, {"answer": "hi"}, ms=42)
assert verify_receipt(receipt)  # recomputes the SHA-256 locally
```

Local receipts follow the AER-1 core members (`id`, `created_at`, `tool`, `provenance_class`, `canonical_bytes`, `output_hash`) and use the provenance class `EXECUTED BY PYDANTICAI-ZAMBO`, naming this package as the recording system. They pass the same hash-recomputation check as any AER-1 receipt.

## Where these receipts run live: Zambo

pydanticai-zambo routes PydanticAI agents through Zambo with a verifiable AER-1 receipt for every call, and it has been downloaded 5,042 times on PyPI. The live home of those receipts is Zambo, the verifiable work layer for AI. Connect any AI to the live endpoint at https://zambo.dev/mcp or add Zambo to your AI in one step at https://zambo.dev/add/

Every tool free once a day, plus 3 continue calls. One free wrap-up handover per day, and pickup is always free. The Day Pass is $0.99 for 24 hours, unlimited.

Install:

    pip install pydanticai-zambo
