How to Use the Cerbos (Access Control) MCP in Pydantic AI

Q: How do I configure Cerbos (Access Control) for Pydantic AI?

Install pydantic-ai-slimmcp. Create an MCPToolset pointing to your Vinkius HTTP endpoint. Assign this toolset to your agent during initialization.

Q: Can the agent delete policies via the server?

Yes, the server includes the deletepolicy tool. Your agent can remove outdated rules by passing the specific policy ID. The framework ensures the ID format is validated before the request fires.

Enforce strict type safety when managing Cerbos (Access Control) policies and permissions with Pydantic AI.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Cerbos (Access Control) MCP to Pydantic AI

Create your Vinkius account to connect Cerbos (Access Control) to Pydantic AI and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Cerbos (Access Control) with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-safe policy management with Pydantic AI

Adding the Cerbos (Access Control) MCP Server to your Pydantic AI setup ensures every policy update matches your exact specifications. When the agent calls `add_policy` or `update_policy`, the framework validates the response against strict Pydantic models. Silent failures do not exist here. If the server returns an unexpected schema after calling `add_schema`, the agent crashes loudly with a validation error. This prevents hallucinated fields from corrupting your access control logic.

Evaluate permissions accurately

Cerbos (Access Control) determines who can access specific resources based on context. Your agent can verify these permissions by executing `authzen_evaluation` or `check_resources`. The model-agnostic nature of Pydantic AI means you can swap between Anthropic or OpenAI without rewriting your authorization logic. Batch processing works exactly the same way. The `authzen_evaluations` tool lets the agent check multiple requests simultaneously. Because the framework prioritizes correctness, you know the returned evaluation array maps perfectly to your requested resources.

Audit and maintain the MCP Server

Visibility into access decisions is non-negotiable, and the Cerbos (Access Control) server exposes this through audit logs. Your AI client can retrieve historical records using the `list_audit_logs` tool. It can then parse these logs into strongly typed objects for further analysis or reporting. Managing the server state is equally straightforward. If a specific rule causes issues, the agent can immediately call `disable_policy`. Once the problem is resolved, it can execute `enable_policy` to restore normal operations.

Setup guide

Set up Cerbos (Access Control) MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "cerbos-access-control-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to Cerbos (Access Control) tools.",
)

result = await agent.run("List recent Cerbos (Access Control) transactions")
print(result.output)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Cerbos. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Cerbos (Access Control) now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Cerbos (Access Control) MCP in Pydantic AI

Install `pydantic-ai-slim[mcp]`. Create an `MCPToolset` pointing to your Vinkius HTTP endpoint. Assign this toolset to your agent during initialization.

The framework supports both Streamable HTTP and SSE transports. You just need to ensure your Vinkius endpoint URL matches the expected protocol for your specific deployment.

Yes, the server includes the `delete_policy` tool. Your agent can remove outdated rules by passing the specific policy ID. The framework ensures the ID format is validated before the request fires.

The framework throws a validation error if the returned schema data does not match the expected Pydantic model. This protects your application from processing malformed data. You can inspect the error trace to see exactly which field failed validation.

The server processes policy configurations, schema definitions, and access evaluation requests. Vinkius manages the authentication layer, requiring only a single endpoint token. This architecture ensures your access control rules remain isolated from your primary application state.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript