Connect your AI agent to MCP

Connect an MCP client so it can call MaxoPerf’s tools. The tools are the same actions you take in the console, and you drive them in plain language.

Faster: install the skill instead

If your agent supports skills (Claude Code, Claude Desktop, and compatible tools), skip the manual config below and go to MaxoPerf agents. One command sets up the connector and teaches the agent MaxoPerf’s workflow.

1. 0:02 Create an API key for your agent
2. 0:34 Connect Claude Code
3. 1:08 Connect Cursor and ChatGPT
4. 1:36 Ask about your runs
5. 2:00 Create and run a test from the agent
6. 2:28 Diagnose a failed run
7. 2:52 Safety

Connect the MaxoPerf MCP server.

Before you start

You need a MaxoPerf API key. Read Get an API key first. It explains that a key grants full-account access and how to use one safely.

Find your endpoint

In the console, open Settings → Connect AI tools (/settings/connect-ai-tools). The page shows your account’s MCP endpoint URL and a config snippet for each supported client, ready to copy. Your endpoint is:

https://app.maxoperf.com/mcp

If the page says “MCP endpoint not configured for this environment,” ask an administrator to configure the MCP server origin for your deployment.

Copy your client's snippet from this page instead of retyping the config below. The snippet already contains your account's real endpoint.

Configure your client

1. Claude Code: run this in your terminal:

   claude mcp add --transport http maxoperf https://app.maxoperf.com/mcp \
     --header "Authorization: Bearer mpak_your_api_key_here"

2. Cursor: add this to .cursor/mcp.json (project or global):

   {
     "mcpServers": {
       "maxoperf": {
         "url": "https://app.maxoperf.com/mcp",
         "headers": {
           "Authorization": "Bearer mpak_your_api_key_here"
         }
       }
     }
   }

3. Claude Desktop / claude.ai: go to Settings → Connectors → Add custom connector, paste the endpoint URL, then paste the API key when asked.

4. ChatGPT: go to Settings → Connectors → Create, paste the endpoint URL, and set the Authorization header to Bearer mpak_your_api_key_here.

Copy your client’s snippet from the console page. It already has your account’s real endpoint, so you only swap in the API key.

Multiple accounts

If your API key’s principal can access more than one account, also set X-Account-Id: <accountId>. See MCP overview.

What the agent can do

Once connected, the agent calls MaxoPerf’s tools to create tests, start runs, poll status, and read KPIs and results. It can do anything the API key is authorized to do. Tools lists the full catalog. The key grants full-account access, so check your tool’s call log before you approve a destructive action (delete run, cancel test, revoke key). If you only want the agent to investigate, connect in read-only mode first.

Choosing what to test

Before you point an agent at a real workload, read Executors. It explains when MaxoPerf runs an API-style load test (JMeter, k6, Taurus, …) and when it runs a full browser test (Playwright, Selenium). That choice changes what the agent should generate.

Where to go next

• MaxoPerf agents: the agent skill that teaches connected agents the full workflow. It is faster than manual config.
• Tools: every tool, its exact arguments, and what it returns.
• Secrets and environments: keep the credentials your agent’s tests use out of test files.