Integrations

Connect Captureze with your existing tools and workflows to automate your monitoring.

AI agents (MCP)

Connect Captureze to Claude, ChatGPT or any agent framework through the Model Context Protocol. Your agent can screenshot a URL and look at the image, watch a page on a schedule, ask what changed since the last capture, pull a Certificate of Capture, and detect a site's cookie banner — 18 tools over the same account and plan limits as the app.

Note: Connecting from Claude signs you in with your ordinary Captureze account — no API key needed. Clients that authenticate with a header need an API key instead, which is available on the Pro plan and during the 7-day trial.

Hosted endpoint

We run the server for you. Point any remote MCP client at this URL — there is nothing to install:

https://mcp.captureze.com/mcp

The endpoint accepts either credential. Clients with a connector dialog and no header field — Claude among them — sign in over OAuth 2.1 against your Captureze account. Everything else authenticates each request with an Authorization: Bearer cap_... header, using a key from Settings → API keys.

Claude (claude.ai, Desktop, mobile)

Add Captureze as a connector — the link opens Claude's Add custom connector dialog with the name and URL already filled in, and you confirm it there. Adding the URL by hand under Settings → Connectors does the same thing. Press Connect, sign in with your Captureze account, and the tools appear in the chat's + menu.

Claude Code

claude mcp add -s user --transport http captureze \
  https://mcp.captureze.com/mcp \
  --header "Authorization: Bearer cap_xxx"

ChatGPT

  1. Go to Settings → Apps → Advanced settings → Developer mode
  2. Add a custom connector with the URL https://mcp.captureze.com/mcp
  3. Set your API key as an Authorization header

Claude Desktop and local agents

Clients that start the server themselves can run it over stdio instead:

{
  "mcpServers": {
    "captureze": {
      "command": "npx",
      "args": ["-y", "@captureze/mcp"],
      "env": { "CAPTUREZE_API_KEY": "cap_xxx" }
    }
  }
}

The server is open source under MIT at github.com/Captureze/mcp. Per-client setup lives in docs/INTEGRATIONS.md, and the launch guide walks through what an agent can actually do with it.

Telegram

Receive instant notifications in Telegram when screenshots are captured or changes are detected.

Setup

  1. Go to Settings → Notifications → Telegram
  2. Click Connect Telegram
  3. You'll be redirected to Telegram to authorize the bot
  4. Send /start to the Captureze bot
  5. Your account will be linked automatically

Bot Commands

  • /start - Link your account
  • /status - Check connection status
  • /stop - Disable notifications

Custom Storage (S3)

Store screenshots in your own S3-compatible storage for full control over your data.

Note: Custom storage is available on the Pro plan.

Supported Providers

  • Amazon S3
  • Cloudflare R2
  • DigitalOcean Spaces
  • MinIO
  • Any S3-compatible storage

Configuration

  1. Go to Settings → Storage
  2. Select S3-Compatible
  3. Enter your storage credentials:
    • Endpoint URL
    • Access Key ID
    • Secret Access Key
    • Bucket Name
    • Region (optional)
  4. Click Test Connection to verify
  5. Save your settings

Webhooks

Integrate with any service that accepts HTTP webhooks. See the Webhooks documentation for details.

Integration Ideas

  • Slack - Send notifications to a Slack channel
  • Discord - Post updates to Discord
  • Zapier - Connect to 5000+ apps
  • Make (Integromat) - Build complex automations
  • n8n - Self-hosted workflow automation
  • Custom scripts - Trigger your own code

REST API

Build custom integrations with our REST API. See the API Reference for complete documentation.

Example Use Cases

  • Programmatically create schedules from your deployment pipeline
  • Trigger screenshots after deployments
  • Build a custom dashboard with your screenshot data
  • Integrate with your CI/CD for visual regression testing

Need a specific integration?

We're always adding new integrations. Let us know what you'd like to see!