Developer Hub

Developer Hub

MCP Server — AI-Native Integration

Sentinel Watch exposes a Model Context Protocol (MCP) server — an open standard that allows AI assistants like Claude to connect directly to your account and interact with live platform data. The MCP server currently provides approximately 20 tools covering cameras, locations, alerts, payouts, reporting, staff management, market benchmarks, and platform configuration.

Connecting Claude Desktop to your Sentinel Watch account via the MCP server takes under five minutes. Once connected, you can query your camera occupancy rates, payout history, alert configuration, and platform benchmarks in plain English — no dashboard required. The MCP server is scoped per API token, so you control exactly which capabilities each connection can access.

  • Camera management — status, schedule, hourly rate, events
  • Location management — configuration and alert routing
  • Alert configuration — per-location and per-camera overrides
  • Attention-check settings — reward, penalty, fail allowance
  • Staff management — list, invite, manage access
  • Payout history — wallet balance and ledger
  • Reporting — events, watch hours, coverage, reward spend
  • Market benchmarks — anonymised platform-wide data
  • Audit log — recent account activity
  • cameras:read / cameras:write
  • alerts:read / alerts:write
  • staff:read / staff:write
  • payouts:read / payouts:write
  • reports:read
  • market:read
  • audit:read
  • events:write
  • wallet:read

REST API — Full Platform Access

The Sentinel Watch REST API provides programmatic access to everything on the platform. Current coverage includes full reporting functions, camera and location management, alert configuration, staff management, payout operations, and account administration. The API uses token-based authentication with granular scope control — identical to the MCP server permission model.

API documentation is available to registered accounts. If you are a current Sentinel Watch customer and need API access, contact us through your dashboard and we will provide your credentials and a link to the full endpoint reference.

AI Chat — Built Into Your Dashboard

Every Sentinel Watch account includes a built-in AI assistant powered by Claude, developed by Anthropic. The assistant is connected to your account via the MCP server, giving it live read access to your cameras, reports, alerts, and market benchmarks. You can ask it questions in plain language, request summaries of recent activity, or get recommendations on rate adjustments based on platform data.

This is not a scripted FAQ bot. It is a capable AI with genuine, permissioned access to your live account — surfacing the same data you would find in your dashboard, but through conversation rather than navigation. Access it from the chat icon in the bottom right of your account dashboard.

What’s Coming Next

The following are actively in development and will be published here as they become available:

  • Public API documentation — full endpoint reference with authentication guides, request/response examples, and error handling
  • Webhooks and event streams — real-time push notifications for camera events, alert triggers, and payout completions
  • SDKs — JavaScript/TypeScript first, with additional language support to follow based on demand
  • Sandbox environment — isolated test credentials for integration development without affecting live data
  • Integration guides — step-by-step setup for common use cases including OttoKit, Zapier, and custom workflows
  • Developer changelog — versioned release notes and deprecation notices

Get API Access

If you are a current Sentinel Watch customer, API credentials and MCP server connection details are available on request. If you are a developer or integration partner interested in building on the platform before it opens publicly, reach out and tell us what you want to build. We are actively working with early integration partners to shape what ships first.

Save to list

Scroll to Top