MCP server — connect any AI agent to MaxoPerf
MaxoPerf runs a Model Context Protocol (MCP) server so an AI agent — in your IDE, in a chat client, or MaxoPerf’s own Loadrigo — can act on your account the same way you would by hand: build a test, start a run, poll it, and read back KPIs and errors, all through natural language.
The endpoint
Section titled “The endpoint”https://app.maxoperf.com/mcpIt speaks MCP’s Streamable HTTP transport, so any client that supports that transport can connect — no MaxoPerf-specific SDK required.
Authentication
Section titled “Authentication”Every request needs two things:
| Header | Value |
|---|---|
Authorization | Bearer mpak_<keyId>_<secret> — a MaxoPerf API key |
X-Account-Id | Your account id, if your API key’s principal has access to more than one account |
An API key grants full-account access — anything you could do signed into the console, an agent holding the key can also do. Treat it like a credential: name keys so you can tell them apart (claude-code-agent, ci-nightly), and revoke one immediately if it leaks.
Read-only mode
Section titled “Read-only mode”A client can restrict itself (or be restricted) to read-only tool calls by sending X-MCP-Read-Only: true on every request. Under read-only mode:
- Read tools (
list_runs,get_test,get_run_results, …) work normally. - Write tools (
start_run,create_test,deploy_virtual_service, …) return aREAD_ONLY_MODEerror instead of executing, so an agent can still see the tool exists — the session just can’t act. cancel_run— being destructive and irreversible — is not even listed in a read-only session’s tool catalog, rather than erroring when called.call_platform_api(the escape hatch) is restricted toGETrequests only.
Use this when you want an agent to investigate and report without any risk of it changing anything.
Response sizing
Section titled “Response sizing”Read tools default to a concise response shape (small, agent-legible field sets) and truncate oversized time series to stay well under typical context budgets. Pass response_format: "detailed" on a call when you genuinely need the full payload.
What’s available
Section titled “What’s available”The server exposes three kinds of things over MCP: tools (the bulk of the surface — see Tools for the full, real catalog), prompts (pre-built multi-step recipes like “run a baseline load test and report the result”), and two read-only resources (maxoperf://openapi and maxoperf://run/{id}) for clients that prefer resources over tool calls.