How to Use the Amazon CloudWatch Log Group MCP in OpenAI Agents SDK

Q: How do I configure the Amazon CloudWatch Log Group MCP Server within the OpenAI Agents SDK?

Install the SDK first using pip, then initialize the MCPServerStreamableHttp class with your Vinkius endpoint. Pass this instance into the mcpservers list when instantiating your Agent to let it auto-discover the filterlogevents tool.

Q: How does tool caching work with the Amazon CloudWatch Log Group MCP Server in OpenAI Agents SDK?

You set cacheToolsList=True in your server parameters to prevent the SDK from repeatedly querying the schema. This speeds up agent response times when calling filterlogevents.

Feed real-time AWS logs directly to your OpenAI Agents SDK pipelines without exposing your entire AWS account.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Amazon CloudWatch Log Group MCP to OpenAI Agents SDK

Create your Vinkius account to connect Amazon CloudWatch Log Group 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 Amazon CloudWatch Log Group with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Target-Lock Log Filtering in OpenAI Agents SDK

The `filter_log_events` tool exposes only one pre-configured CloudWatch Log Group to your OpenAI Agents SDK. You don't have to worry about your agent wandering off into other AWS services or reading logs it shouldn't touch. This single-purpose MCP Server locks the agent's scope to exactly one log group, keeping your production environment safe. Because the LogGroupName is hardcoded on the server side, the agent cannot manipulate the target group, preventing any accidental or malicious lateral movement across your AWS infrastructure.

Guardrail-Protected AWS Debugging

The `filter_log_events` tool works directly with OpenAI's built-in guardrails to validate log queries before they run. If your agent tries to pull an absurdly large range of log records, your system catches it. This prevents runaway API bills and keeps token usage under control. You get full tracing on the OpenAI dashboard for every log search query, showing exactly what log patterns were checked.

Low-Latency Tool Discovery

The `filter_log_events` tool registers instantly via the streamable HTTP MCP transport. By setting `cacheToolsList=True` in your Python setup, your production agents discover the tool with zero startup overhead. This setup avoids slow API handshakes when your agent needs to inspect a fresh error. Your system gets immediate access to the logs, allowing for rapid diagnostics and fast handoffs between specialized agents.

Setup guide

Set up Amazon CloudWatch Log Group 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 Amazon CloudWatch Log Group tools at runtime.
3

Create your Agent

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

Run the agent

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

Install the SDK first using pip, then initialize the `MCPServerStreamableHttp` class with your Vinkius endpoint. Pass this instance into the `mcp_servers` list when instantiating your `Agent` to let it auto-discover the `filter_log_events` tool.

No, the agent cannot target a different group. The server strictly hardcodes the log group name, so the `filter_log_events` tool only queries that specific resource. Even if the agent attempts to pass a different group name, the server ignores it.

You set `cacheToolsList=True` in your server parameters to prevent the SDK from repeatedly querying the schema. This speeds up agent response times when calling `filter_log_events`.

You should use OpenAI's built-in guardrails to filter out highly sensitive patterns before they reach the agent. Since the server only exposes `filter_log_events`, you can also restrict the types of queries the agent is allowed to construct.

Your raw log events are processed in an ephemeral, zero-trust V8 sandbox. Only the specific log records matching your query are transmitted to the OpenAI Agents SDK, ensuring no log data is ever persisted or stored on the Vinkius platform.

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