How to Use the Diese MCP in Pydantic AI

Q: How do I register the Diese MCP Server in Pydantic AI?

Initialize an MCPToolset with your server's HTTP endpoint. Pass this toolset into your Agent instance to automatically expose tools like listerpprojects and listoverdueerptasks to your model.

Q: What happens if Diese returns unexpected data to Pydantic AI?

The framework raises a ValidationError immediately at runtime. What's the status of your database? If it drifts, this error saves you from writing corrupted records.

Q: Can I use SSE transport to connect Diese to Pydantic AI?

Yes, the unified MCPToolset supports both SSE and streamable HTTP transports. You specify the connection protocol when setting up the toolset in your Python application.

Q: How do I handle contact lookups with strict validation?

Your agent calls listerpcontacts and parses the returned JSON into a predefined Pydantic schema. This ensures that every contact's email, phone number, and organization field are fully validated before use.

Q: How is my Diese resource planning data protected when using this MCP Server?

All data fetched via listresourceplanning is processed in memory on your host and isolated via Vinkius's zero-trust sandboxes. Your sensitive scheduling and ERP project data never touches persistent storage on our end, ensuring strict compliance with your internal security policies.

Run type-safe ERP workflows with Pydantic AI and Diese to guarantee your agent never processes malformed invoice or project data.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Diese MCP to Pydantic AI

Create your Vinkius account to connect Diese 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 Diese with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Runtime type safety for Diese project tasks

The `diese-mcp` MCP Server integrates with Pydantic AI to enforce strict schema validation on every ERP payload. When your agent calls `list_project_tasks` or `get_project_details`, the incoming JSON is validated against strict Pydantic models before your agent can use it. If the API schema changes unexpectedly, the system throws a runtime error instead of silently passing corrupted data. This prevents common LLM errors where models hallucinate fields that do not exist in your ERP. Your agent gets clean, typed Python objects for tracking tasks and resource allocations. You can write your logic with total confidence that the fields returned by `list_resource_planning` match your expected types.

Validate business invoices using this MCP Server

Your Pydantic AI agents validate financial records at runtime using `list_sales_invoices` to ensure your billing pipelines remain accurate. When retrieving invoices via `list_sales_invoices` or expenses through `list_business_expenses`, the framework checks every field against your data models. Any missing payment status or malformed currency string triggers an immediate validation exception. This level of strictness is critical when automating financial operations. You can build agents that process expense approvals without worrying about a model misinterpreting a null value as zero. The system halts execution immediately if the data structure doesn't match your exact specifications.

Model-agnostic project search with strict outputs

This MCP setup allows you to switch underlying LLMs while querying `search_projects_by_name` without rewriting your ERP integration code. Switch between Anthropic, OpenAI, or local models without breaking your validation schemas. The model-agnostic nature of the toolset ensures consistent behavior across different provider APIs. You manage your connections using the unified `MCPToolset` class, which handles the underlying SSE or HTTP transport. If you need to verify account limits with `get_account_metadata`, the tool returns a strongly-typed metadata object. This keeps your application code clean and decoupled from specific model provider quirks.

Setup guide

Set up Diese 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": {
        "diese-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

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

result = await agent.run("List recent Diese 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 Diese. 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 Diese 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 Diese MCP in Pydantic AI

Initialize an `MCPToolset` with your server's HTTP endpoint. Pass this toolset into your `Agent` instance to automatically expose tools like `list_erp_projects` and `list_overdue_erp_tasks` to your model.

The framework raises a `ValidationError` immediately at runtime. What's the status of your database? If it drifts, this error saves you from writing corrupted records.

Yes, the unified `MCPToolset` supports both SSE and streamable HTTP transports. You specify the connection protocol when setting up the toolset in your Python application.

Your agent calls `list_erp_contacts` and parses the returned JSON into a predefined Pydantic schema. This ensures that every contact's email, phone number, and organization field are fully validated before use.

All data fetched via `list_resource_planning` is processed in memory on your host and isolated via Vinkius's zero-trust sandboxes. Your sensitive scheduling and ERP project data never touches persistent storage on our end, ensuring strict compliance with your internal security policies.

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