What this skill does
Filling out the same PDF form over and over — client intake forms, permit applications, vendor contracts — is exactly the kind of task that should be a five-second ask instead of a ten-minute chore. This skill reads a blank PDF form, maps its fields, and fills it out from either a plain-language brief or a structured data source (a Sheet row, a JSON object), producing a completed PDF ready to send.
How it works
- Field mapping — the skill inspects the PDF's form fields (or, for non-fillable PDFs, its visual layout) and builds a map of what each field expects.
- Data resolution — it matches your supplied data (free text, or a row of structured values) to the mapped fields, flagging anything it can't confidently resolve rather than guessing.
- Fill and export — it writes the values into the form fields and exports a completed PDF, leaving the original template untouched.
Requirements
- Works on Claude AI (Claude Code, or claude.ai with code execution enabled) — PDF field manipulation runs via a Python library invoked through code execution.
- Fillable (AcroForm) PDFs work most reliably; scanned/flat PDFs are supported via layout-based field detection but should be reviewed before sending.
Setup
- Unzip the package into your Skills directory, keeping the folder structure intact.
- Provide the data source you want to fill from (plain text brief, CSV row, or JSON object) alongside the blank PDF.
- Try it with a sample form using the included worked example.
What you get
- A blank PDF form automatically mapped, field by field, without you manually identifying each one
- Your data matched into the right fields from a plain brief, a spreadsheet row, or a JSON object
- Anything it can't confidently resolve flagged instead of silently guessed
- A completed, ready-to-send PDF with the original template left untouched