How to Use the DocBreach MCP in OpenAI Agents SDK

Q: How does DocBreach prevent token bloat in the OpenAI Agents SDK?

docs.read strips away all HTML boilerplate, navigation sidebars, and tracking scripts before the payload hits your OpenAI Agents SDK pipeline. You receive clean, structured markdown, which keeps your context window clear and minimizes tracing costs on your OpenAI dashboard.

Feed clean API specs straight to your OpenAI Agents SDK pipelines without spinning up heavy headless browsers.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect DocBreach MCP to OpenAI Agents SDK

Create your Vinkius account to connect DocBreach 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 DocBreach with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Target API Specs Directly with OpenAI Agents SDK

`docs.extract` pulls raw, structured endpoint schemas from OpenAPI, Swagger, or Postman URLs directly into your agent's context. Your OpenAI system bypasses slow browser rendering entirely, feeding clean JSON structures straight into the agent's function-calling loop. This setup prevents the agent from hallucinating parameters by verifying actual payload formats before running tools. You get immediate validation against real API specifications, which works alongside the native guardrails in the SDK to block invalid API calls.

Map Unknown API Structures on the Fly

`docs.map` crawls a domain to construct a clean table of contents that your specialized agents inspect during runtime handoffs. Instead of dumping raw HTML into the context, the tool returns organized sections to let your planning agent decide which specific sub-pages to read next. Once the map is ready, your agent calls `docs.read` on specific URLs to pull down clean, LLM-ready markdown. This targeting keeps your token usage low and ensures that your OpenAI dashboard traces show clean documentation reads rather than massive, unparsed web pages.

Find Missing Endpoints via Targeted Search

`docs.search` queries specific documentation sites using the mandatory site parameter to resolve developer queries without leaving the execution loop. Your agent queries local indexes for authentication headers or specific error codes, avoiding generic web searches that lead to stale information. For broader discovery, `docs.discover` runs descriptive queries to locate the correct API reference homepages. The OpenAI Agents SDK caches this MCP tool list using the cache parameter, keeping execution speeds fast while your agent hunts down the right integration details.

Setup guide

Set up DocBreach 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 DocBreach tools at runtime.
3

Create your Agent

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

Run the agent

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

`docs.read` strips away all HTML boilerplate, navigation sidebars, and tracking scripts before the payload hits your OpenAI Agents SDK pipeline. You receive clean, structured markdown, which keeps your context window clear and minimizes tracing costs on your OpenAI dashboard.

Yes, you register the MCP server once in your streamable HTTP configuration, and all agents in the run pool access the tools. The planning agent can use `docs.map` to find the correct endpoints, then hand off the specific URL to an execution agent running `docs.read`.

No, the server handles all documentation discovery and extraction without requiring third-party search keys or browser automation platforms. You only need the single Vinkius endpoint token inside your OpenAI Agents SDK streamable HTTP parameters.

The tool expects a single, highly descriptive query rather than rapid, repetitive search calls. Refine your search string with specific terms like 'Stripe API webhooks' to get the best result on the first try and avoid rate limits.

Your target documentation URLs and raw API specs are processed in an ephemeral V8 Isolate sandbox that destroys all session data immediately after extraction. No raw text or schema payloads are written to persistent storage, keeping your internal API structures completely private.

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