MCP●Watch

Know before your users do
when your MCP server breaks.

Uptime and schema-contract monitoring built for the Model Context Protocol. We run the real MCP handshake every minute, snapshot your tool schemas, and alert you the moment a tool goes down — or a breaking change ships that silently breaks every agent calling you.

Join the waitlist — first 50 servers monitored free during beta.
$ mcpwatch https://api.yourserver.com/mcp
● UP    https://api.yourserver.com/mcp
  server: catalog-api  protocol: 2025-06-18  latency: 41ms  tools: 12
  ⚠ 2 BREAKING contract change(s) since 2026-06-14T09:12:00Z:
    - tool removed: get_item
    - search: new required param "locale"
  1 non-breaking change(s):
    - search: optional param added "limit"
→ alert sent to #eng-alerts (Slack) + webhook

What it catches that uptime monitors can't

Real MCP handshake

Not a dumb HTTP ping. We initialize + tools/list like a real client, so "200 OK but broken" never slips through.

Schema-contract diffing

We snapshot every tool's input schema and flag breaking changes — removed tools, new required params, type changes — before they break your callers.

Alerts that matter

Email, webhook, or Slack the instant something goes down or a contract breaks. Latency & uptime history included.

Pricing

FREE
$0/mo
  • 1 server
  • 15-min checks
  • Public index listing
  • Downtime email alerts
MOST POPULAR
$19/mo
  • 10 servers
  • 1-min checks
  • Schema-change alerts
  • Email + webhook + Slack
  • 30-day history
TEAM
$49/mo
  • 50 servers
  • 30-sec checks
  • Private dashboards
  • Multiple alert channels
  • 1-year history

The MCP Reliability Index

We publicly track the uptime and reliability of well-known MCP servers — a free, neutral leaderboard for the ecosystem. Is your server on it? Claim & monitor it →