How to Use the Webflow MCP in AutoGen

Q: Can Webflow and AutoGen help me manage CMS items?

Yes. You can have one agent propose createcollectionitem while another challenges the input data payload for completeness. The system won't proceed until consensus is reached on the content structure.

Automate Webflow deployments using AutoGen. Agents debate the best path for publishing your site.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Webflow MCP to AutoGen

Create your Vinkius account to connect Webflow to AutoGen 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 Webflow with AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Governing Site Publishing Decisions

Don't just publish—let agents argue about it. You can set up a consensus group where one agent checks permissions using `get_user`, another validates content completeness via `list_collection_items`, and a third calls `publish_site` only when all parties agree. This forces deliberation before deployment. The system doesn't just execute; multiple agents debate the risk profile of running `publish_site`, leading to safer, more thoughtful outcomes.

Auditing and Creating Collections

Need to validate content? Have one agent list all current collections (`list_collections`) while another attempts to create new ones using `create_collection_item`. They will debate if the schema is correct or if a conflict exists. This consensus-driven approach means you don't just write code; you build a review board. The agents argue over whether an item needs updating (`update_collection_item`) versus simply creating a new one.

Mapping and Listing Webflow Sites

The group can be tasked with mapping all web properties. One agent uses `list_sites` to get the full list, while another uses `get_site` on each entry to validate its current status and domain ownership. They debate which sites are ready for review or if they need a manual check. It’s about convergence on a complete, validated inventory of your Webflow assets.

Setup guide

Set up Webflow 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 Webflow 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="Webflow_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

result = await agent.run("List recent Webflow 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="Webflow_assistant",
        model_client=OpenAIChatCompletionClient(model="gpt-4o"),
        workbench=workbench,
    )

    result = await agent.run("List recent Webflow 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 Webflow. 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 Webflow 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 Webflow MCP in AutoGen

You'll set up multiple agents: one acts as the developer calling tools like `get_site`, while another acts as the QA reviewer, challenging those results. They debate until a final plan is agreed upon.

Yes. You can have one agent propose `create_collection_item` while another challenges the input data payload for completeness. The system won't proceed until consensus is reached on the content structure.

This is perfect for governance. Agents can debate: 'Is this site ready?' Agent A calls `list_pages` to check scope; Agent B calls `publish_site`. The final decision requires agreement from both sides.

The agents can debate the permissions required. One agent uses `get_user` to check current credentials, while another challenges whether those credentials are sufficient for the intended action.

It handles site structure (domains), content inventory (collections/pages), and user permissions. The agents use these inputs to debate the best course of action for the entire system.

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