How to Use the TfL MCP in AutoGen

Q: How does AutoGen handle TfL journey planning?

The agents start by calling getjourney for initial plans. Then, one agent might challenge the plan by running getplacesearch to see if a better-located stop exists, leading to a revised route consensus.

Q: What if I use AutoGen to plan a bike route with TfL?

An 'Optimization' agent uses getbikepoints to find available docks. A second agent confirms the location and accessibility using getstoppointdetails. They debate whether the dock is convenient enough for the user.

Run consensus-driven decision making with AutoGen over TfL data.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers — Included with Plan

Connect TfL MCP to AutoGen

Create your Vinkius account to connect TfL to AutoGen — we handle the hosting, security, and runtime updates so you don't have to. No server setup required.

GDPR Included with Plan

Setup TfL with AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity Vinkius runs on Gemini

Gemini

Grok

Debating Optimal Journeys

You can set up a multi-agent system where one agent uses `get_journey` to suggest the fastest path, and another challenges that suggestion using `get_road_disruptions`. They debate whether the time savings are worth the risk of localized road incidents. The final decision is not just the 'best' answer; it’s the consensus reached after competing perspectives argue over the data.

Cross-Checking Service Reliability

A Performance Agent might call `get_arrivals` to confirm immediate service times. Meanwhile, a Safety Agent calls `get_line_status` to check for systemic issues like 'Part Suspended.' The agents must then reconcile these two conflicting data points (real-time vs. macro status) before giving the user an answer. This simulates human deliberation when the source data is ambiguous.

Vetting Vehicle Compliance

One agent checks if a vehicle registration is ULEZ compliant using `get_vehicle_details`. A second, 'Logistics' agent then uses `get_road_disruptions` to determine if the specific route segment has any temporary weight or size restrictions. They debate if the initial compliance check even matters given the roadworks. It forces a complex vetting process over simple API lookups.

Setup guide

Set up TfL MCP in AutoGen

Prerequisites

Python 3.10+ installed
autogen-ext[mcp] package
Active Vinkius subscription with a valid endpoint token

1

Install AutoGen with MCP
Run pip install "autogen-ext[mcp]" autogen-agentchat. The MCP extension includes mcp_server_tools for stateless tool access.
2

Fetch tools from the MCP
Call mcp_server_tools(SseServerParams(url=...)) with your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Run your agent
Pass the tools to AssistantAgent and call agent.run(). The agent invokes TfL tools and returns structured results.

agent.py

from autogen_ext.tools.mcp import SseServerParams, mcp_server_tools
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient

server_params = SseServerParams(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

tools = await mcp_server_tools(server_params)

agent = AssistantAgent(
    name="TfL_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

result = await agent.run("List recent TfL data")
print(result.messages[-1].content)

Get your connection token →

Prerequisites

Python 3.10+ installed
autogen-ext[mcp] + autogen-agentchat
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Same packages as above. McpWorkbench is ideal when your agent needs stateful sessions across multiple tool calls.
2

Use McpWorkbench as context manager
Wrap your agent in async with McpWorkbench(...) to maintain shared state and resources. The workbench manages the full MCP session lifecycle.
3

Run with workbench
Pass workbench=workbench to your agent. State is preserved across multiple tool calls within the same session.

agent.py

from autogen_ext.tools.mcp import McpWorkbench, SseServerParams
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient

server_params = SseServerParams(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

async with McpWorkbench(server_params) as workbench:
    agent = AssistantAgent(
        name="TfL_assistant",
        model_client=OpenAIChatCompletionClient(model="gpt-4o"),
        workbench=workbench,
    )

    result = await agent.run("List recent TfL data")
    print(result.messages[-1].content)

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 TfL. 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 TfL 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 TfL MCP in AutoGen

The agents start by calling `get_journey` for initial plans. Then, one agent might challenge the plan by running `get_place_search` to see if a better-located stop exists, leading to a revised route consensus.

Absolutely. You can assign an 'Operations' agent the task of calling `get_line_status`. This agent reports back on severity (e.g., Minor Delays) and disruption details, which other agents then build their recommendations around.

An 'Optimization' agent uses `get_bike_points` to find available docks. A second agent confirms the location and accessibility using `get_stop_point_details`. They debate whether the dock is convenient enough for the user.

You'll build a 'Verification' agent whose only job is to cross-reference `get_road_disruptions` against any proposed route. This makes the system highly skeptical and forces it to acknowledge potential closures.

The server touches public infrastructure data, including vehicle registrations (`get_vehicle_details`), stop point coordinates, and operational status. The agents only process these non-personal system metrics.

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