How to Use the Actionstep MCP in Pydantic AI

Q: How do I initialize the Actionstep MCP in Pydantic AI?

Install the slim package with pip install "pydantic-ai-slimmcp". Create an MCPToolset pointing to your Vinkius HTTP endpoint, then provide it to the toolsets argument when building your Agent.

Q: What happens if Actionstep returns an unexpected matter type to Pydantic AI?

The agent fails loudly. When listactiontypes returns data that breaks your Pydantic schema, the framework throws a runtime validation error. This prevents silent failures in your legal workflows.

Q: Can Pydantic AI read my Actionstep file notes?

Your agent executes listmatternotes to retrieve the text. The framework validates the response payload, ensuring every note has the correct timestamp and author fields before passing the text to your chosen model.

Type-safe Actionstep MCP integration for Pydantic AI. Validate legal tasks, billable hours, and case matters at runtime.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Actionstep MCP to Pydantic AI

Create your Vinkius account to connect Actionstep 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 Actionstep with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Actionstep MCP Server for strict schemas

The `get_matter_details` tool returns complex case metadata from your practice management system. Your Pydantic AI agent maps this raw JSON directly into strict Python models. If the API returns a string where an integer should be, the framework throws a validation error immediately. You catch bad data before your LLM hallucinates a response based on corrupted case files.

Manage contacts with guaranteed types

Your agent uses `create_contact` to push new clients into your database. The framework ensures every required field exists and matches your exact type definitions before the request even leaves your server. Reading works the same way. When calling `list_contacts`, the returned address book gets parsed through your Pydantic schemas. You build reliable intake workflows that never crash due to a missing email key.

Audit time entries safely

Connect the `list_time_entries` tool to pull billing records for specific attorneys or matters. The agent extracts the exact hours and descriptions logged in the system. Because Pydantic AI is model-agnostic, you swap between local models for simple data extraction and Anthropic for complex billing audits. The `MCPToolset` standardizes the transport, so your code stays identical regardless of the underlying LLM.

Setup guide

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

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

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

Install the slim package with `pip install "pydantic-ai-slim[mcp]"`. Create an `MCPToolset` pointing to your Vinkius HTTP endpoint, then provide it to the `toolsets` argument when building your Agent.

The agent fails loudly. When `list_action_types` returns data that breaks your Pydantic schema, the framework throws a runtime validation error. This prevents silent failures in your legal workflows.

Your agent executes `list_matter_notes` to retrieve the text. The framework validates the response payload, ensuring every note has the correct timestamp and author fields before passing the text to your chosen model.

Yes. The framework abstracts the LLM completely. You run a local model to parse `list_legal_tasks` and extract deadlines, keeping the entire operation on your own hardware while still using the standardized toolset.

Vinkius completely isolates the connection. Your billable time, matter details, and client contacts pass through a zero-trust proxy. The MCP server requires a single authentication token and immediately destroys the sandbox environment after the HTTP request finishes.

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