Install the MaxoPerf skill
The MaxoPerf skill is a small bundle for skill-aware AI agents (Claude Code, Claude Desktop, and compatible tools). It sets up the MCP connector and gives the agent the MaxoPerf workflow: how to pick a test type, where secrets go, and what a healthy run looks like. Without it, you would explain all of that in a long prompt.
Install
Section titled “Install”-
Get a MaxoPerf API key first. See Get an API key. Keep it ready, because the installer asks for it.
-
Run the installer from a terminal where your agent (Claude Code, etc.) is available:
Terminal window npx @maxoperf/agent-skill install -
Paste your API key when prompted. The installer stores it the same way a manual MCP config would. It never prints or logs the key.
-
Restart or reload your agent so it loads the new skill and MCP connector.
You can also start from the console. Settings → Connect AI tools (/settings/connect-ai-tools) shows the same install command with a copy button and links to this page.
First run
Section titled “First run”After you install, ask your agent for something concrete: “create a load test against my staging checkout endpoint” or “show me the last run for project X.” With the skill, the agent can:
- Pick a test type. See Choosing an executor for how it decides between API-style and browser-style engines.
- Keep credentials out of the test files it generates. See Secrets and environments.
- Explain each MCP tool call in plain language before it runs anything destructive.
Safety
Section titled “Safety”The rule is the same as for manual MCP setup: an API key grants full-account access. Review your agent’s tool-call log before you approve create, run or delete actions, especially the first few times. See Get an API key for how to scope and rotate keys.
Uninstall or update
Section titled “Uninstall or update”You can re-run npx @maxoperf/agent-skill install at any time. It is idempotent and updates the skill in place. To remove the skill, use your agent’s normal skill-removal flow (for example, Claude Code’s /skills management). If you no longer need the API key, revoke it in Settings → API keys.
Where to go next
Section titled “Where to go next”- Connect AI tools: the manual MCP setup that this skill automates.
- Choosing an executor: the engines the agent chooses between when it creates a test.
- Secrets and environments: how the agent should handle credentials in generated tests.