How to Use the Percy MCP in OpenAI Agents SDK

Q: How do I add the Percy toolset to my OpenAI Agents SDK project?

Use MCPServerStreamableHttp to connect the endpoint, pass it to the Agent constructor, and let auto-discovery expose tools like listprojects automatically.

Q: Can an agent built with OpenAI Agents SDK automatically approve failing Percy builds?

Yes, the agent can call approvebuild or approvesnapshot. You should use the SDK's built-in guardrails to restrict these write actions to specific branches.

Q: What happens if a visual comparison fails during an agent run?

The agent uses listcomparisons to catch the high diff percentage. It then halts the pipeline or hands off the task to a human-in-the-loop agent for triage.

Run automated visual checks and approve Percy snapshot builds directly inside your production OpenAI Agents SDK pipelines.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers — Included with Plan

Connect Percy MCP to OpenAI Agents SDK

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

Ask AI about this MCP

ChatGPT

Claude

Perplexity Vinkius runs on Gemini

Gemini

Grok

Safely automate visual QA gates with OpenAI Agents SDK

Look, visual regressions are a silent killer. This MCP Server exposes Percy's visual testing platform so you catch them early. Your agent uses `list_builds` to track incoming test runs and `get_build_details` to inspect the exact count of unreviewed visual changes. By registering the server within the `mcp_servers` list, your agent autonomously detects UI drift on every commit. The built-in guardrails in the SDK prevent unauthorized approvals by forcing verification steps before running `approve_build`.

Inspect pixel-level regressions on demand

This MCP Server queries Percy snapshot details directly from your codebase. The agent uses `list_snapshots` and `get_snapshot_details` to isolate broken layouts. Check the diff. Fix it or ship it. When a visual test fails, the agent triggers a handoff to a specialized triage agent. This secondary agent calls `list_comparisons` to evaluate the exact diff percentage across different browser targets before requesting human review.

Approve visual baselines dynamically

This MCP Server manages visual baseline approvals directly from your pipeline. It provides direct endpoints to modify baseline images using `approve_snapshot` for individual components, or it executes `approve_build` to unblock the build pipeline. You configure this by adding the MCP Server HTTP endpoint to your agent constructor using `MCPServerStreamableHttp`. Enabling `cacheToolsList=True` ensures your production agents resolve these Percy tools instantly without API roundtrip lag.

Setup guide

Set up Percy 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 Percy tools at runtime.
3

Create your Agent

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

Run the agent

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

Use `MCPServerStreamableHttp` to connect the endpoint, pass it to the `Agent` constructor, and let auto-discovery expose tools like `list_projects` automatically.

Yes, the agent can call `approve_build` or `approve_snapshot`. You should use the SDK's built-in guardrails to restrict these write actions to specific branches.

The SDK automatically queries the endpoint on startup, registers all ten visual testing tools, and caches them for fast execution when you set `cacheToolsList=True`.

The agent uses `list_comparisons` to catch the high diff percentage. It then halts the pipeline or hands off the task to a human-in-the-loop agent for triage.

This server only processes metadata, snapshot states, and project slugs. All pixel-diff images and DOM snapshots are transmitted securely via the Vinkius sandboxed transport directly to Percy, never touching external logs.

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