How to Use the Holded MCP in Pydantic AI

Q: How do I configure the Holded MCP Server in Pydantic AI?

Use the unified MCPToolset class with your Vinkius HTTP endpoint. Do not use the deprecated HTTP server classes. Pass the toolset directly to your agent.

Q: What happens if Holded returns malformed data to my Pydantic AI agent?

The framework catches it immediately. If getprojectdetails returns an unexpected string instead of a boolean, the agent fails with a clear validation error.

Q: How does my Pydantic AI agent find specific Holded task IDs?

The agent runs listalltasks to pull the active roster. It validates the response array and extracts the exact task ID needed for further processing.

Guarantee type-safe Holded API responses in Pydantic AI. Your agent validates every invoice and contact against strict schemas.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Holded MCP to Pydantic AI

Create your Vinkius account to connect Holded 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 Holded with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Validate Holded billing data

The `list_invoices` and `get_invoice_details` tools return billing data that Pydantic AI validates at runtime. If the API spits back an unexpected currency format, the framework throws a loud validation error. You never have to deal with silent corruption. Your agent knows exactly what fields exist in the invoice response because the MCP Server enforces the schema. It processes financial data with absolute correctness.

Manage stock with Pydantic AI

Running `update_product_stock` requires a specific JSON body containing the warehouse ID and quantity. Pydantic AI forces your agent to construct this payload perfectly before it ever hits the network. You pair this with `get_product_details` to verify the change. The agent reads the updated stock count, validates the integer response, and continues its workflow. You get predictable, safe write operations.

Query client records safely

Your agent calls `list_contacts` and `get_contact_details` to pull client profiles into your typed Python code. It handles leads, suppliers, and clients through the same validated pipeline. Configure this by installing `pydantic-ai-slim[mcp]`. Pass the Vinkius URL to `MCPToolset` and inject it into your agent's toolsets array. The framework handles the Streamable HTTP transport and maps the Holded tools to your models automatically.

Setup guide

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

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

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

Use the unified `MCPToolset` class with your Vinkius HTTP endpoint. Do not use the deprecated HTTP server classes. Pass the toolset directly to your agent.

The framework catches it immediately. If `get_project_details` returns an unexpected string instead of a boolean, the agent fails with a clear validation error.

Yes. Pydantic AI is model-agnostic. You can connect local LLMs or cloud providers to the MCP Server and they will all use the same validated tools.

The agent runs `list_all_tasks` to pull the active roster. It validates the response array and extracts the exact task ID needed for further processing.

Vinkius handles authentication securely without exposing your API keys to the client. When your agent pulls warehouse stock levels via `get_product_details`, the request executes inside an ephemeral, sandboxed environment that destroys itself immediately after.

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