# mcp-server-mojaq

Read-only [MCP](https://modelcontextprotocol.io) server for [MOJAQ](https://mojaq.com) — ask Claude, Claude Code or Cursor about your EU-hosted app's **deploys, errors, logs, uptime, incidents and root-cause hints**, in natural language.

Because MOJAQ keeps all of these in one place, your AI assistant can *correlate* across them — "errors spiked 4 minutes after deploy `v2.4.0` while flag `new-checkout` was on" — right inside your editor. It is **strictly read-only**: no tool can change or delete anything.

## Setup

You need a MOJAQ API key (from [mojaq.com/app/api-keys](https://mojaq.com/app/api-keys)).

**Claude Desktop / Claude Code / Cursor** — add to your MCP config:

```json
{
  "mcpServers": {
    "mojaq": {
      "command": "npx",
      "args": ["-y", "mcp-server-mojaq"],
      "env": { "MOJAQ_API_KEY": "mq_your_key_here" }
    }
  }
}
```

Then ask your assistant things like:

- *"What broke in the last hour?"*
- *"Show me the errors on /checkout and tail the payment-service logs for timeouts."*
- *"What did the last deploy change, and did error rates go up after it?"*
- *"Is anything down right now?"*

## Tools

| Tool | What it does |
|------|--------------|
| `project_summary` | Monitors + status, open incidents, recent deploys, root-cause hints, log volume |
| `get_timeline` | One chronological feed of deploys, incidents, flag changes and hints |
| `explain_recent_regression` | Timeline for the last N hours, to reason about a recent regression |
| `query_logs` | Search logs by text, level, service and time window |
| `get_deploys` | Recent deployments |
| `uptime_status` | Uptime monitors and their status |
| `search_errors` | Recent error-tracking events |

## Config

| Env | Required | Default |
|-----|----------|---------|
| `MOJAQ_API_KEY` | yes | — |
| `MOJAQ_API_URL` | no | `https://api.mojaq.com` |

Read-only. MIT licensed. EU-hosted (Helsinki).