Extract booking rows from Interhome-style arrival, check-in, or check-out PDF lists and write them to a CSV file. Use this skill whenever the user mentions an Interhome arrival list, a guest check-in/check-out PDF, owner bookings marked as OB, or wants stay rows turned into CSV from one of these PDFs, even if they do not explicitly ask for a "skill" or say "Interhome" in the final instruction.
Use this skill for Interhome-style booking list PDFs where each row represents a stay and the goal is structured CSV output.
Write a CSV file with this exact header row:
arrival_date,leaving_date,invoice_no,guest,owner_booking
Use UTF-8 encoding. Dates must be normalized to YYYY-MM-DD.
This skill is for PDFs that look like an Interhome arrival or check-in/check-out list:
Check-in / Check-out column with two dates in the same cellInvoice No columnContact column that may contain name, email, and phoneContactIf the PDF does not match that structure, stop and say why instead of inventing rows.
arrival_date: first date from the Check-in / Check-out cell, converted from DD/MM/YY to YYYY-MM-DDleaving_date: second date from the same cell, converted the same wayinvoice_no: keep as a string exactly as shownguest: the first non-empty line from Contact; discard email addresses and phone numbersowner_booking: true when the country / owner-booking cell is marked OB, otherwise falseContact cell. In that case, keep guest empty.OB badge may be graphical instead of extractable text. The bundled script handles this by checking the rendered owner-booking cell for the blue Interhome badge.owner_booking=true when the badge is detected.Identify the input PDF from the user's prompt or attached files.
Choose an output CSV path in the working directory unless the user gave one explicitly.
Run the bundled extractor:
python3 scripts/extract_arrival_list.py /path/to/input.pdf --output /path/to/output.csv
If the environment already has a project-local Python environment, prefer that interpreter. If pdfplumber is missing, install it in the active local environment before running the script.
Hand back the CSV file path and keep the response brief.