How to Use the Directus MCP in OpenAI Agents SDK

Q: How do I connect Directus to the OpenAI Agents SDK?

Install the package via pip, then initialize the MCPServerStreamableHttp client with your endpoint. Pass this server instance into your Agent constructor using the mcpservers list. The agent automatically discovers all ten tools.

Q: How does performance scaling work when using Directus with the OpenAI Agents SDK?

Set cacheToolsList=True when configuring your streamable HTTP server. This keeps tool schemas cached in memory, preventing redundant network roundtrips every time the agent initializes or switches contexts.

Q: What is the best way to handle agent handoffs for database operations?

Create a read-only agent that uses getsingleitem and a separate write-authorized agent for running patchcmsrecord. Use the SDK's native handoff mechanism to transfer the conversation only when a database update is required.

Build production-grade OpenAI Agents SDK systems that read, write, and audit your Directus SQL data via this MCP Server.

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 OpenAI Agents SDK

Create your Vinkius account to connect Directus to OpenAI Agents SDK 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 OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Inspect databases directly with this MCP Server

Give your OpenAI agents the ability to read your entire database structure without manual mapping. By exposing `list_schema_collections` and `list_collection_fields`, the agent inspects PostgreSQL schemas on the fly, finding exactly which tables and columns exist before running operations. This prevents database errors during complex agent runs. Your agent queries specific columns using `list_collection_items` to verify data types, ensuring it has the correct schema context before trying to write or update any records.

Production-grade record mutation

Let your agents handle data entry and record updates directly within your CMS. Using `create_cms_record` and `patch_cms_record`, the agent writes JSON payloads to your tables or updates existing entries by their unique IDs. Because this runs inside the OpenAI Agents SDK, you can wrap these mutation tools in custom Python guardrails. This setup ensures that destructive tools like `wipe_cms_record` require explicit human validation before any database row actually gets deleted.

User and asset tracking

Keep track of who is doing what and where your files are stored. The agent uses `list_directus_users` to verify active admin accounts and matches media storage assets using `list_directus_files`. This makes it easy to run automated audits or generate reports. If an agent needs to check a specific log or trace a database row UUID, it calls `get_single_item` to pull the exact record details instantly.

Setup guide

Set up Directus MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Directus tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Directus tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Directus tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Directus Agent",
            instructions="You have access to Directus tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

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 OpenAI Agents SDK

Install the package via pip, then initialize the `MCPServerStreamableHttp` client with your endpoint. Pass this server instance into your Agent constructor using the `mcp_servers` list. The agent automatically discovers all ten tools.

Yes. While the server exposes all tools by default, you can filter the tool list in your Python code before passing them to the Agent constructor. This is highly recommended for disabling destructive tools like `wipe_cms_record` in production.

Set `cacheToolsList=True` when configuring your streamable HTTP server. This keeps tool schemas cached in memory, preventing redundant network roundtrips every time the agent initializes or switches contexts.

Create a read-only agent that uses `get_single_item` and a separate write-authorized agent for running `patch_cms_record`. Use the SDK's native handoff mechanism to transfer the conversation only when a database update is required.

Your database credentials and raw SQL rows remain strictly within the Vinkius V8 sandbox. The server only transmits the specific JSON payloads requested by your agent. No database credentials or schema definitions are stored outside your local environment.

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