Help › For AI agents
Connect your AI agent
Give your AI assistant UserAlertX and it can text you, ask your approval, read your agent inbox, email you summaries and send video messages. It can only ever reach you, never anyone else.
The UserAlertX connector (MCP)
Most AI assistants can add an MCP connector. Use:
- Server URL:
https://useralertx.com/mcp - Header:
Authorization: Bearer YOUR_API_KEY(the key from your account screen)
What your agent can do with it
send_sms_to_me: text you an alert.request_approval/confirm_approval: ask your OK first, with a code only you get.read_inbox: read your AgentInboX: an email address your AI agent can read (always marked untrusted).email_me_summary: a UserSummaryX: email me the gist summary to your own email.send_avatar_message: an AvatarMsgX: text into a talking video message video message by email.text_my_group/list_my_groups: text one of your own MemberAlertX: group texts groups, like "text my group Smith Family: dinner at 6".
Your assistant
- Claude Code: see Text me when Claude Code is done or waiting.
- Grok Bot: add the connector (https://useralertx.com/grokbotmcp.html). Can't add connectors? Have it email your alert address: Send an alert by email.
- ChatGPT: Business, Enterprise and Edu plans can add the connector. Other plans use the Personal Alert Link on your account screen (https://useralertx.com/chatgptintegration.html).
- Home Assistant: a short REST notifier in configuration.yaml (https://useralertx.com/homeassistant.html).
- Anything else that can make a web request or send an email: Send an alert from code (API) or Send an alert by email.
TipTell your agent in its instructions: "When you finish a long task, text me a one-line summary via UserAlertX. Before deleting files or spending money, ask my approval via UserAlertX first."
Related topics
Text me when Claude Code is done or waitingStart a long Claude Code task, walk away, and get a text when it's done or waiting for you: a permission prompt, a question, a finished run.TeXt Approval: approve by textPut uxapprove in front of anything that should need your OK first: a backup cleanup, a deploy, a script an AI agent runs.Send an alert from code (API)One POST request sends an alert.