manage-third-party-services

Manage third-party services

Preflight, read first. If you cannot see any uptimerobot:* MCP tools in your tool list, invoke the uptimerobot:setup skill before doing anything else. Do not tell the user the MCP is misconfigured, setup's Step 0 detects the common case (server connected, tools loaded after session start) and resolves it without re-keying.

A third-party service tracks the public status of an outside provider (a cloud, a payment processor, a SaaS) and alerts the account when the provider reports an incident. It is a monitor of its own type, but it lives on its own tools:

• list-monitors does not return third-party services, and monitorsCount / monitorLimit from get-user do not count them. Use list-third-party-services.
• The generic tools (get-monitor-details, update-monitor, delete-monitor) are not for them. Use the *-third-party-service tools below.
• Two kinds of ID are in play. A provider ID identifies a provider in the catalog and is used only to find and start tracking it. A monitor ID identifies the tracked service on the account and is used for everything after that. Both are plain numbers. Never use a provider name, status page URL or slug as an ID, and never guess either kind.

Use this skill when the user says "track AWS status", "alert me when Stripe has an incident", "what third-party services am I tracking", "only watch these components", or "stop tracking X".

Step 1: Find the provider

{"search":"stripe","limit":10}

Call search-third-party-providers. search matches a provider or component name; limit is 1-50. Each row has the provider id, name, statusPageUrl, state (operational, minor, major or maintenance), the matchedComponent the search hit (or null), and isTracked.

• Pick the provider by its numeric id. If several rows plausibly match, show the user the names and status page URLs and ask which one.
• If isTracked is true, the account already tracks it. Find the monitor ID with list-third-party-services instead of creating again.