TypeScript · Claude SDK · long memory · escalation guardrail
Alagà · Guest AI Agent
Knows the whole stay, acts with real tools, escalates everything an agent should never freelance.
status live run + 4 real appsrole designed + built end to end
The raw run, then the actual receipts it produced across four apps. 35 seconds.
What it solves
A guest agent that re-asks questions is worse than a form. One that improvises on a refund or a 2am lockout is a liability. So the memory spans the stay, and the escalation check runs both before the model answers and again after, which means a smooth, confident answer cannot argue its way past the rule.
Who it serves
Short-stay and hospitality operators who want a guest-facing agent without handing it judgment calls.
Who it is for
Written to a founding-engineer bar: useful and safe at the same time.
How it works
- One guest messageThe agent loads the whole stay from memory, not just the last line, so it never re-asks what it knows.
- The guardrail runs firstAnything about money, safety, or legal is force-escalated in code before the model can answer, and backstopped after.
- The model actsFor everything else it uses real tools that do work, not 'I'll let the host know'.
- It fires a real workflowThe decision hits an n8n 'Guest Ops' flow that writes to Slack, Gmail, Google Calendar, and Airtable in one run.
Built with
TypeScript Claude SDK node:sqlite memory tool-use loop n8n dispatch Slack · Gmail · Calendar · Airtable