# maxoperf

Control MaxoPerf load and browser tests through an AI agent, then inspect run metrics and diagnostics. Access requires a MaxoPerf API key.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 1
- Author: MaxoPerf
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/MaxoPerf/mcp
- Updated: 2026-07-22
- HTML: https://affpapa.org/en/nevibecode/mcp/maxoperf

## What it does

The hosted server connects an AI agent to MaxoPerf. It can create tests, upload their files, start runs on managed cloud runners, and retrieve results. Available data includes throughput, latency percentiles, error rates, logs, and runner health.

The toolset covers projects, tests, runs, API discovery, and failure analysis. Text recipes guide tasks such as comparing p95 across runs or investigating a failed run. Nothing needs to run locally, but requests require `MAXOPERF_API_KEY` for platform authentication. In `read-only` mode, write tools are hidden.

## Requirements
- Remote server: `https://app.maxoperf.com/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http maxoperf https://app.maxoperf.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add maxoperf --url https://app.maxoperf.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "maxoperf": {
            "url": "https://app.maxoperf.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "maxoperf": {
            "type": "http",
            "url": "https://app.maxoperf.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http maxoperf https://app.maxoperf.com/mcp
```

## When to use it
- create a load test and launch a run
- compare p95 across two runs
- inspect errors and logs after a failed run
- check whether runners reached the target load
- review latency and throughput metrics

## Example prompts
- Run a load test against my API and show me its results.
- Compare p95 for two runs and explain where latency increased.
- Investigate the failed run by checking its errors, logs, and achieved runner load.

## Similar
- [load-testing-specialist](https://affpapa.org/en/nevibecode/agents/load-testing-specialist.md) — subagent · 986 installs
- [k6-perf-test-website](https://affpapa.org/en/nevibecode/skills/k6-perf-test-website.md) — skill · 3400 installs
- [performance-testing](https://affpapa.org/en/nevibecode/skills/performance-testing.md) — skill · 1100 installs
- [151-java-performance-jmeter](https://affpapa.org/en/nevibecode/skills/151-java-performance-jmeter.md) — skill · 349 installs
- [performance-engineer](https://affpapa.org/en/nevibecode/skills/performance-engineer.md) — skill · 918 installs
- [setup-load-testing](https://affpapa.org/en/nevibecode/commands/setup-load-testing.md) — command · 235 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md