OpenClaw to send reminders from booking files
Felix Krause built a travel assistant around booking confirmations rather than asking an agent to buy every flight and hotel. A scheduled job finds upcoming reservations and converts them into structured Markdown files in a shared travel repository.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Read the prepared booking Markdown files and send useful Telegram reminders for upcoming travel. Flag missing reservations, checkout details and packing needs; keep booking records read-only.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Felix Krause
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The integration delivers travel-context messages and reminders from existing bookings; no new paid booking or quantified saving is reported.
- Human involvement
- Krause manually labeled the source emails; the travel bot had read-only access to generated reservation files.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Label travel confirmations in Fastmail for the scheduled parsing job.
- Store structured reservations in the shared Markdown travel folder.
- Let the read-only travel bot send timely Telegram reminders from those records.
Felix Krause built a travel assistant around booking confirmations rather than asking an agent to buy every flight and hotel. A scheduled job finds upcoming reservations and converts them into structured Markdown files in a shared travel repository.
The email boundary is deliberately narrow. Krause manually applies a Travel label in Fastmail, and the parsing job only reads those tagged messages. Fastmail's read-only token supports that separation. A dedicated travel bot can read the generated files and use selected information tools, including maps, routing and flight APIs.
Its recurring checks turn the stored reservations into timely messages for a Telegram travel group. Krause describes travel-day overviews, hotel-checkout information, missing-booking reminders and details such as gates or seating near boarding. A packing list he wrote once supplies personal context for checklists matched to the type of trip.
The source reports this as an integration he set up and shows the resulting message workflow. It does not quantify saved money or claim that the bot completed new paid bookings. The specific pattern a reader can copy is separating ingestion from the travel assistant: a limited email job creates the reservation files, and the travel bot works from that prepared information. Human tagging remains part of deciding which messages enter the system.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: krausefx.com · Felix Krause
Frequently asked questions
Did it read every email?
No. The parsing job only read messages Krause manually tagged Travel.
Why separate the travel bot from ingestion?
The dedicated bot could work from generated Markdown bookings with read-only access and selected information skills.
Where did packing preferences come from?
Krause wrote a reusable list of items he might pack, which supplied context for trip-specific checklists.
More details
About this account
OpenClaw checked prepared booking files and selected travel tools to send contextual Telegram reminders.
Firsthand source opened and read. Suggested prompt written by us from the reported workflow; no original task prompt is quoted.
Credit: Felix Krause
Source published
Added to Agents In Use
Source checked
Tools and topics
OpenClaw · Fastmail · Markdown · Telegram · Travel APIs
Booking confirmations · Read-only travel bot · Packing context
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.