Start here
What is Escanor
Escanor connects your code host, cloud, databases, monitoring and chat, with an assistant, an MCP server for coding agents, and desktop and phone apps.
Last updated
On this page
Escanor is an operational layer that connects the tools your software runs on: your code host, your cloud, your databases, your monitoring and your chat. You can see all of it in one dashboard, ask an assistant to investigate or change things, give the same tools to your coding agents, and keep an eye on everything from your phone.
The pieces
| Piece | What it is | Where |
|---|---|---|
| Dashboard | The web app: integrations, projects, deployments, incidents, activity, monitoring, the assistant, settings and billing. | escanor.in in any browser |
| Assistant | Chat that can look things up and act on your connected tools. Changes follow the configured permissions and approval policy. | Dashboard, desktop app and phone app |
| Escanor MCP | One MCP server that gives your coding agents (Claude Code and others) five tools that reach every integration you connected. | Your agent's MCP settings |
| Desktop app | Escanor with your own computer as the server: local models, Docker, an editor, files, and pairing with your phone. | Linux, Windows and macOS |
| Phone app | Chat, voice, approvals, your computers and your machines in your pocket. | Android (APK), iPhone (sideload) |
| Machines | Servers you own running the Escanor agent, so you can drive Claude Code on them from anywhere. | Connect a VM |
How they fit together
- You sign in once and connect your integrations. Escanor keeps those connections signed in for you; if one ever expires, you reconnect it.
- The dashboard and the assistant read from those integrations: what is deployed, what is failing, who changed what.
- Your agents reach the same integrations through the Escanor MCP with a revocable token, so you do not paste secrets into prompts.
- The desktop app and the phone app are the same account in different places. A phone pairs with a computer using a short code and then controls it, with approvals for anything risky.
What it is good for
- Seeing production at a glance: deploys, failing checks, open incidents and recent changes across tools.
- Asking "why did this break?" and getting an answer grounded in your actual deployments and logs.
- Letting a coding agent open a pull request, check a deployment or restart a service without handing it your keys.
- Approving or rejecting what an agent wants to do from your phone while you are away from your desk.
Where to go next
- New here? Read the Quick start.
- Want to know what can be connected? See Integrations.
- Curious about safety? Read Security and privacy.
Need help? Contact support with a redacted error and the affected version.