How to Use the Oracle Vector DB MCP in OpenAI Agents SDK

Q: Can OpenAI Agents SDK automatically discover indexes in Oracle Vector DB?

Yes, the agent calls listvectorindexes to discover HNSW and IVF vector indexes. This lets the agent inspect the existing vector layout before executing queries.

Q: How do I prevent raw SQL execution errors when using Oracle Vector DB with OpenAI Agents SDK?

Your agent should call describetable to check the exact columns and data types first. Always enforce the FETCH FIRST 100 ROWS ONLY clause inside executesqlquery to keep payloads small.

Run native vector searches directly from your OpenAI Agents SDK pipelines into Oracle 23ai databases.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers — Included with Plan

Connect Oracle Vector DB MCP to OpenAI Agents SDK

Create your Vinkius account to connect Oracle Vector DB to OpenAI Agents SDK — we handle the hosting, security, and runtime updates so you don't have to. No server setup required.

GDPR Included with Plan

Setup Oracle Vector DB with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Native Vector Queries via OpenAI Agents SDK

The `vector_search` tool runs native vector similarity calculations directly inside your Oracle 23ai database using native VECTOR_DISTANCE functions. Your Python agents execute semantic searches against high-dimensional vector columns and retrieve ordered nearest neighbors without pulling raw data into memory first. By using this MCP Server with your agent pipelines, you bypass the need for external vector databases. The agent determines when to call `vector_search` or fallback to traditional queries, keeping your database operations unified under one connection pool.

Database Schema and Index Discovery

The `list_vector_indexes` tool pulls active HNSW or IVF index metadata so your agent understands the indexing structure before executing queries. This prevents the agent from running unindexed, expensive vector scans over massive tables, protecting your database CPU from spiking. Combine this with `describe_table` and `list_tables` to give your agent a clear map of the database schema. The agent checks explicit column types, including native vector types, to build exact SQL statements without guessing table names or column definitions.

Direct SQL Execution with Guardrails

The `execute_sql_query` tool runs raw SQL statements against your Oracle runtime using ORDS when standard vector searches are not enough. Because large payloads degrade agent performance, this tool requires a strict FETCH FIRST 100 ROWS ONLY limit to keep payloads tiny and fast. You can monitor these raw database executions directly inside your OpenAI dashboard. The agent combines `table_stats` with execution plans to verify row counts and optimizer stats before running heavy queries, keeping your production database stable.

Setup guide

Set up Oracle Vector DB 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 Oracle Vector DB tools at runtime.
3

Create your Agent

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

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Oracle Vector DB 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="Oracle Vector DB Agent",
            instructions="You have access to Oracle Vector DB 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 Oracle Database. 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 Oracle Vector DB 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 Oracle Vector DB MCP in OpenAI Agents SDK

You register the MCP Server using the MCPServerStreamableHttp client in Python. The SDK auto-discovers tools like `vector_search` and exposes them directly to your agents for runtime execution.

Yes, the agent calls `list_vector_indexes` to discover HNSW and IVF vector indexes. This lets the agent inspect the existing vector layout before executing queries.

Your agent should call `describe_table` to check the exact columns and data types first. Always enforce the FETCH FIRST 100 ROWS ONLY clause inside `execute_sql_query` to keep payloads small.

Vinkius hosts the MCP Server and handles the connection pool to your Oracle database on the backend. Your Python code only needs to maintain a single HTTP stream connection to the server endpoint.

The MCP Server runs in a zero-trust V8 sandbox on Vinkius, meaning your raw database credentials and schema data never touch external logs. Only the specific tool outputs, like table names from `list_tables`, are passed to your agent over encrypted HTTP.

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