How to Use the Liveblocks (Collaborative) MCP in AutoGen

Q: How do AutoGen agents collaborate on document states using Liveblocks (Collaborative)?

One agent reads the state using getydoc, while another proposes changes. The writer agent then uses this MCP server to update the room via updateydoc or patchstorage once they reach consensus.

Q: Can I restrict which Liveblocks (Collaborative) tools an AutoGen agent can use with this MCP Server?

Yes. You can register only specific tools, like listthreads or createthread, to your agent configuration. This prevents agents from executing administrative tools like deleteroom without explicit permission.

Q: How does AutoGen handle user authorization with Liveblocks (Collaborative)?

Your agents use authorizeuser and identifyuser to generate secure, short-lived tokens. This allows the agents to verify user permissions before granting access to collaborative workspaces.

Q: Can agents broadcast real-time events to active users?

Yes. The broadcastevent tool allows your agents to send JSON payloads directly to active clients in a room. This is useful for triggering UI animations or sending instant alerts without saving data to disk.

Let your AutoGen agents debate and coordinate real-time document updates or user permissions.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Liveblocks (Collaborative) MCP to AutoGen

Create your Vinkius account to connect Liveblocks (Collaborative) 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 Liveblocks (Collaborative) with AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Resolve document state conflicts in AutoGen debates

The `update_ydoc` tool allows your AutoGen agents to push binary updates to a shared document after reaching consensus. In an AutoGen group, one agent might propose an edit, while a reviewer agent checks it for errors. Once they agree, the designated AutoGen writer agent sends the change, ensuring only clean updates reach your live users. To keep the AutoGen agents aligned, they use `get_ydoc` to read the current state of the document. If they detect a conflict during their debate, they run `patch_storage` to fix specific keys in the shared JSON tree. This collaborative workflow prevents AutoGen agents from overwriting human edits or fighting over document states.

Manage user permissions via collaborative MCP Server tools

The `authorize_user` tool requests secure access tokens for users based on AutoGen agent discussions. For instance, an AutoGen security agent can review a user's request, while an admin agent executes the tool to grant specific permissions. This multi-agent verification process prevents unauthorized changes to your rooms. Room metadata is kept up to date using `update_room` to adjust permissions or tags based on AutoGen decisions. The AutoGen agents use `identify_user` to verify identities before modifying any access levels. This ensures your collaborative spaces remain secure without requiring manual admin intervention in AutoGen.

Coordinate team communication and presence in AutoGen

The `create_thread` tool lets your AutoGen agents start new discussion threads to alert human collaborators about important changes. If an AutoGen health agent detects an issue, it posts a comment to a room to notify the team. The AutoGen agents track these discussions using `list_threads` to stay updated on human feedback. Ephemeral presence is managed via `set_presence`, allowing AutoGen agents to show their active status on the user interface. They use `list_active_users` to see which humans are online before broadcasting events with `broadcast_event`. This coordination ensures AutoGen agents only ping users who are actually active in the room.

Setup guide

Set up Liveblocks (Collaborative) 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 Liveblocks (Collaborative) 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="Liveblocks (Collaborative)_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

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

    result = await agent.run("List recent Liveblocks (Collaborative) 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 Liveblocks. 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 Liveblocks (Collaborative) 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 Liveblocks (Collaborative) MCP in AutoGen

One agent reads the state using `get_ydoc`, while another proposes changes. The writer agent then uses this MCP server to update the room via `update_ydoc` or `patch_storage` once they reach consensus.

Yes. You can register only specific tools, like `list_threads` or `create_thread`, to your agent configuration. This prevents agents from executing administrative tools like `delete_room` without explicit permission.

Your agents use `authorize_user` and `identify_user` to generate secure, short-lived tokens. This allows the agents to verify user permissions before granting access to collaborative workspaces.

Yes. The `broadcast_event` tool allows your agents to send JSON payloads directly to active clients in a room. This is useful for triggering UI animations or sending instant alerts without saving data to disk.

All API calls to modify room metadata or delete rooms are executed inside zero-trust sandboxes. This MCP server ensures your actual room configurations and access tokens are never logged or stored.

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