How to Use the KEGG MCP in Pydantic AI

Q: What happens if the KEGG API changes and my Pydantic AI agent gets an unexpected response?

Your agent will raise a ValidationError immediately. This is the core benefit—it fails loudly at the source of the problem, so you know exactly where the data mismatch occurred, rather than chasing bugs downstream.

Q: Is it complicated to set up Pydantic AI with this KEGG MCP server?

No, it's straightforward. You just instantiate the MCPToolset with the Vinkius server URL and pass it to your agent. The library handles the tool discovery and validation from there.

Get validated, type-safe genomic data in your Python app by connecting Pydantic AI to the KEGG database.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect KEGG MCP to Pydantic AI

Create your Vinkius account to connect KEGG 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 KEGG with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Never Trust Bad API Data Again

When your agent calls `kegg_get` for a pathway file or `kegg_list` for a list of organisms, Pydantic AI is working behind the scenes. It takes the raw text response from the KEGG API and forces it to match a Pydantic model you define. If the data is malformed, incomplete, or just not what you expected, your code gets a `ValidationError`. You don't get silent data corruption, you get an immediate, loud failure that you can catch and handle.

Swap LLMs, Keep Your Tools

Build your entire biological query logic using KEGG tools like `kegg_find` and `kegg_link`. Because Pydantic AI is model-agnostic, that logic works the same way everywhere. You can prototype with an OpenAI model, then switch to a local Llama model for production without changing a single line of your tool-using code. The framework handles the model-specific parts, so your KEGG workflow is completely portable.

Type-Safe ID Conversions with this MCP Server

Mapping identifiers between databases is a common source of bugs. The `kegg_conv` tool does the conversion, but Pydantic AI makes it safe. By defining the expected output format in a Pydantic model, you guarantee that a call to `kegg_conv` returns exactly the string format you need. This prevents subtle bugs that come from an agent hallucinating an ID format or the API returning something unexpected.

Setup guide

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

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

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

It validates every response from the KEGG MCP server against a Pydantic model at runtime. If the data from a tool like `kegg_get` doesn't match the schema you defined, it raises a `ValidationError` instead of passing bad data to your application.

Yes. Pydantic AI is model-agnostic. You can connect it to OpenAI, Anthropic, Google, or a model running on your own machine, and your KEGG tool definitions will work without any changes.

Your agent will raise a `ValidationError` immediately. This is the core benefit—it fails loudly at the source of the problem, so you know exactly where the data mismatch occurred, rather than chasing bugs downstream.

No, it's straightforward. You just instantiate the `MCPToolset` with the Vinkius server URL and pass it to your agent. The library handles the tool discovery and validation from there.

Your data, such as entry identifiers and organism names, passes through a Vinkius ephemeral instance that's wiped clean after the request. Pydantic AI itself processes the data in-memory for validation and doesn't store it, keeping your queries private.

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