HostTracker MCP server

Connect an AI assistant to HostTracker over the Model Context Protocol and let it operate your monitoring account in conversation: run a live check from 300+ global locations, see what is down, create or pause a monitor, schedule a maintenance window, review incidents, manage who gets alerted, wire a webhook, publish a status page update.

Endpoint   https://mcp.host-tracker.com/mcp
Transport  streamable HTTP
Auth       Authorization: Bearer <your HostTracker API token>

This repository is the public face of that hosted server: the connection metadata (server.json), the per-client setup guide (CLIENT.md), and the security policy (SECURITY.md). The server itself is a hosted service, so there is nothing here to install or run.

Availability: the full tool surface described below is rolling out. If your client lists only the three instant-check tools after connecting, it reached the previous version of the endpoint; reconnect after the rollout completes.

Connect in two minutes

1. Mint a token. Go to Integrations → API and create a token with the scopes you want the assistant to have (start with check and monitor:read).

2. Point your client at the endpoint with that token in an Authorization header. Configuration blocks for every common client are below.

3. Reload the client so it discovers the tools, then ask in plain language: "is example.com up right now, checked from Europe and Asia?", "which of my monitors are down?", "pause the staging monitor until tomorrow".

Claude Code

.mcp.json in your project (or ~/.claude.json for a user-wide connector):

{
  "mcpServers": {
    "hosttracker": {
      "type": "http",
      "url": "https://mcp.host-tracker.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_HOSTTRACKER_API_TOKEN" }
    }
  }
}

Or from the command line:

claude mcp add --transport http hosttracker https://mcp.host-tracker.com/mcp \
  --header "Authorization: Bearer YOUR_HOSTTRACKER_API_TOKEN"

Claude Desktop

The "Add connector" dialog accepts OAuth connectors only, and this server authenticates with a bearer token, so Desktop connects through the mcp-remote bridge. Edit claude_desktop_config.json:

{
  "mcpServers": {
    "hosttracker": {
      "command": "npx",
      "args": [
        "-y", "mcp-remote", "https://mcp.host-tracker.com/mcp",
        "--header", "Authorization:${HT_AUTH}"
      ],
      "env": { "HT_AUTH": "Bearer YOUR_HOSTTRACKER_API_TOKEN" }
    }
  }
}

The ${HT_AUTH} indirection is deliberate: some mcp-remote builds split an argument on its first space, which breaks a literal Authorization: Bearer .... Node.js 18 or newer is required.

Cursor

~/.cursor/mcp.json for every project, or .cursor/mcp.json for one:

{
  "mcpServers": {
    "hosttracker": {
      "url": "https://mcp.host-tracker.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_HOSTTRACKER_API_TOKEN" }
    }
  }
}

VS Code (GitHub Copilot agent mode)

.vscode/mcp.json in the workspace. The inputs block keeps the token out of the file, prompting for it once and storing it in the editor's secret storage:

{
  "inputs": [
    {
      "type": "promptString",
      "id": "ht-token",
      "description": "HostTracker API token",
      "password": true
    }
  ],
  "servers": {
    "hosttracker": {
      "type": "http",
      "url": "https://mcp.host-tracker.com/mcp",
      "headers": { "Authorization": "Bearer ${input:ht-token}" }
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "hosttracker": {
      "serverUrl": "https://mcp.host-tracker.com/mcp",
      "headers": { "Authorization": "Bearer YOUR_HOSTTRACKER_API_TOKEN" }
    }
  }
}

Docker (stdio bridge)

For a client that only speaks stdio, or to try the server without installing Node.js, the repository's Dockerfile packages the mcp-remote bridge in front of the hosted endpoint:

docker build -t hosttracker-mcp https://github.com/HostTracker/mcp.git
docker run -i --rm -e HT_TOKEN=YOUR_HOSTTRACKER_API_TOKEN hosttracker-mcp

Point the client at that docker run command as a stdio server. Nothing of the service runs in the container; it forwards to https://mcp.host-tracker.com/mcp under your token.

ChatGPT and other clients

Any client that speaks MCP over streamable HTTP and can send a static header works. Give it the endpoint https://mcp.host-tracker.com/mcp and the header Authorization: Bearer YOUR_HOSTTRACKER_API_TOKEN. Where a client's connector form offers an API-key or custom-header authentication mode, the token goes there; where it offers OAuth only, use the mcp-remote bridge shown under Claude Desktop.

A generic, dependency-free bridge for anything that can only launch a command:

npx -y mcp-remote https://mcp.host-tracker.com/mcp --header "Authorization:${HT_AUTH}"

Longer walkthroughs, verification commands and troubleshooting live in CLIENT.md.

Related MCP server: uptrack-mcp

What the assistant can do

The server exposes the HostTracker v2 REST API as MCP tools. Every list tool takes limit (maximum 50) and cursor and returns the next cursor; every timestamp is Unix seconds in both directions; ids are opaque strings.

Three behaviours worth knowing before the first call:

• Bulk operations validate first. A bulk tool returns a validation report; the write needs an explicit second call with submit=true. Bulk deletion additionally needs confirmed=true and the count the validation pass reported, so a selection that drifted in between is refused.

• Bulk operations and reports are asynchronous. They answer with a job id; poll it with wait_for_job.

• Deleting anything is not undoable, so it always takes two calls. A delete tool's first call removes nothing - it returns the resource so the assistant can show you what is about to go - and only a repeat call with confirmed=true deletes. The API returns a receipt listing what was removed.

Authentication and scopes

Mint tokens at Integrations → API. Scopes are per family with :read and :write leaves that do not imply each other; a bare family name satisfies every leaf under it.

Grant the narrowest set that covers the work. There is never a reason to grant account:write: the server refuses every write under /account regardless of what the token allows. If a call comes back refused, ask the assistant to run get_account_quota, which reports the scopes the token actually carries.

Limits and quota

• The endpoint rate-limits each client IP on /mcp. A limited response carries Retry-After, which the server passes through as a value; it never sleeps or retries on your behalf, so the assistant decides what to do.

• Your API plan quota is enforced against your own token, exactly as it is for direct REST calls. get_account_usage and get_account_quota report where you stand. Details in the errors and limits guide.

• Application failures (no token, wrong scope, quota exhausted, invalid input) come back as ordinary tool results with an actionable message rather than a protocol error.

Safety

The server is stateless and stores nothing: every call is forwarded to the API under your own token, and all ownership, quota and rate enforcement happens there. On top of that it refuses, at the server, regardless of the token: any write under /account, and anything touching payments, plans, passwords, login or the minting of API tokens. Content that a checked target controls is wrapped in a fenced, length-capped block before it reaches the model, so a hostile target cannot inject instructions into your assistant. Full policy in SECURITY.md.

Links

• MCP server overview

• Mint an API token

• API v2 reference and guides

• All integrations

• Official clients: JavaScript · Python · Go · .NET · CLI · OpenAPI description · GitHub Action

• Support: ht2support@host-tracker.com

License

The contents of this repository (documentation and metadata) are released under the MIT license. The hosted MCP server and the HostTracker service itself are proprietary.