# UptimeBolt MCP Server

AI-powered infrastructure monitoring — query monitor health, incidents, AI predictions, root cause analysis, deploy safety checks, and executive summaries via MCP.

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add uptimebolt/mcp-server

# Browse available tools
npx -y @smithery/cli@latest tool list uptimebolt/mcp-server

# Get full schema for a tool
npx -y @smithery/cli@latest tool get uptimebolt/mcp-server get_service_status

# Call a tool
npx -y @smithery/cli@latest tool call uptimebolt/mcp-server get_service_status '{}'
```

## Direct MCP Connection

Endpoint: `https://mcp-server--uptimebolt.run.tools`

**Required config:**
- `x-api-key` (header) — Your UptimeBolt API key (starts with uptb_)

## Tools (10)

- `get_service_status` — Get the current health status of a service or all services. Returns health score, monitor status, and active incidents.
- `get_predictions` — Get active AI predictions for monitors or services. Shows predicted problems with confidence levels.
- `get_incidents` — Get incidents with optional filters. Includes root cause analysis if available.
- `get_monitor_health` — Get detailed health information for a specific monitor including response time, uptime, and active predictions.
- `get_deployments` — Get recent deployments and their correlation with incidents. Shows which deploys potentially caused issues.
- `run_root_cause_analysis` — Run an AI-powered root cause analysis for an incident or service. Analyzes dependencies, cascading failures, and deploy…
- `is_safe_to_deploy` — Check if it's safe to deploy right now based on current service health, active predictions, and recent incidents. Usefu…
- `get_executive_summary` — Get an executive summary of infrastructure health for a time period. Ideal for daily standups, weekly reports, or statu…
- `get_monitors` — List all monitors with optional filtering by status or type. Returns name, URL, operational status, response time, and …
- `get_monitor_metrics` — Get detailed metrics summary for a specific monitor including response time stats, uptime percentage, and error breakdo…

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get uptimebolt/mcp-server <tool-name>
```

---

License: MIT