Run3 analyze pdf calendar

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill run3_analyze_pdf_calendar

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Parse the calendar PDF to identify existing appointments, flexible blue slots, and the 15-minute grid coordinate mapping.

1.3 KB, as published. Nobody here has run it

Open /root/calendar.pdf using a PDF parsing library (e.g., pdfminer or PyMuPDF).
Timezone Identification: Search the text content of the PDF for a timezone string (e.g., "PST", "EST", "UTC") to ensure all scheduled times match the document's context.
Coordinate Mapping:
- Identify horizontal lines across the calendar timeline.
- Determine the vertical distance (Δy) between any two adjacent horizontal lines. This distance represents exactly 15 minutes.
- Establish a reference Y-coordinate for a known time (e.g., the top line of the workday).
Appointment Extraction:
- Extract all rectangular path objects/shapes from the PDF.
- Determine the start and end times of each block by mapping its top and bottom Y-coordinates to the timeline grid.
- Color Detection: Inspect the color metadata for each block. Use the RGB color space.
  - Blocks with an RGB value of (0, 0, 1) (pure blue) are identified as flexible/low-priority slots.
  - Blocks of other colors (e.g., grey, red) are identified as busy/fixed slots.
Output: A structured timeline of busy and flexible blocks, the coordinate-to-time ratio, and the detected timezone.