How to Use the Directus MCP in AutoGen

Q: How do AutoGen agents resolve conflicts when writing to Directus?

The writer agent attempts updateitem and passes any API errors back to the group. A developer agent then analyzes the error and coordinates with the writer to adjust the payload. This back-and-forth negotiation continues until the write succeeds or is safely aborted.

Q: Can I restrict which Directus tools an AutoGen agent can use?

Yes, you can register a limited subset of tools like listitems for reader agents while reserving createitem for writer agents. This MCP setup makes it easy to isolate write permissions and prevent unauthorized model actions by enforcing separation of duties.

Q: How does AutoGen handle schema mapping for complex Directus databases?

The agents use listfields to inspect the available columns and build a shared understanding of your database model. They then debate the best way to query the data before calling searchitems. This collaborative planning prevents hallucinated field names.

Let your AutoGen agents debate and coordinate database updates in Directus through secure multi-agent workflows.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Directus MCP to AutoGen

Create your Vinkius account to connect Directus 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 Directus with AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Coordinate database writes via AutoGen agents

Your primary assistant uses `create_item` and `update_item` to modify database records, but only after a second agent verifies the input parameters. This setup allows you to build multi-agent validation loops where one agent drafts a content update and another checks it for compliance before execution. If the reviewer agent flags an error, the writer agent calls `delete_item` or corrects the payload in a follow-up turn. This cooperative debugging happens entirely in the background before any changes commit to production.

Inspect schemas dynamically using the MCP Server

The agents run `list_collections` and `list_fields` over this MCP connection to negotiate how to map user requests to your SQL database. One agent might suggest a table structure, while a DBA agent verifies if those fields actually exist in your Directus instance. This collaborative schema discovery prevents broken queries and API errors during execution. The group converges on the correct database fields before making any read or write calls.

Audit system activity through collaborative agents

Your monitoring agent uses this MCP Server to query `list_activity` and `list_users` to track who is making changes to your database. This lets you build conversational audit systems where you can ask your agent group who modified a specific record and why. By checking `list_roles`, the security agent ensures the current user has the correct permissions to trigger the requested action. This adds a layer of conversational safety to your automated database workflows.

Setup guide

Set up Directus 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 Directus 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="Directus_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

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

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

The writer agent attempts `update_item` and passes any API errors back to the group. A developer agent then analyzes the error and coordinates with the writer to adjust the payload. This back-and-forth negotiation continues until the write succeeds or is safely aborted.

Yes, you can register a limited subset of tools like `list_items` for reader agents while reserving `create_item` for writer agents. This MCP setup makes it easy to isolate write permissions and prevent unauthorized model actions by enforcing separation of duties.

The agents use `list_fields` to inspect the available columns and build a shared understanding of your database model. They then debate the best way to query the data before calling `search_items`. This collaborative planning prevents hallucinated field names.

The agent stops execution and notifies the user that the Directus API token is invalid or expired. You will need to check your environment variables to ensure the connection has the correct permissions.

No, your database activity, role lists, and user profiles are processed entirely in memory on your host. The MCP Server acts as an ephemeral proxy that does not store or forward your database transactions to external logs. All traffic remains strictly between your local runner and your Directus instance.

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