Run1 python pdf form filling

Skill cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/court-form-filling/run1_python-pdf-form-filling

[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 run1_python-pdf-form-filling

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

Use this skill when you need to programmatically fill PDF form fields using Python. Covers inspecting field names and writing values to fillable PDFs.

SKILL.md

4.1 KB, as published. Nobody here has run it

Filling PDF Forms with Python

Recommended Libraries

Option 1: PyPDF2 / PyPDF (pypdf)

from pypdf import PdfReader, PdfWriter

reader = PdfReader("input.pdf")
writer = PdfWriter()

# Clone the original
writer.append(reader)

# Inspect field names
fields = reader.get_fields()
for field_name, field_obj in fields.items():
    print(f"Field: {field_name}, Type: {field_obj.get('/FT')}, Value: {field_obj.get('/V')}")

Filling fields:

from pypdf import PdfReader, PdfWriter

reader = PdfReader("sc100-blank.pdf")
writer = PdfWriter()
writer.append(reader)

# Fill fields by name
fill_data = {
    "FieldName1": "Value1",
    "FieldName2": "Value2",
}

writer.update_page_form_field_values(writer.pages[0], fill_data)
# For multi-page forms, repeat for each page index

with open("sc100-filled.pdf", "wb") as f:
    writer.write(f)

Option 2: pdfrw

import pdfrw

template = pdfrw.PdfReader("sc100-blank.pdf")

# Iterate annotations to find fields
for page in template.pages:
    annotations = page.get('/Annots')
    if annotations:
        for annot in annotations:
            field_name = annot.get('/T')
            field_type = annot.get('/FT')
            print(f"Field: {field_name}, Type: {field_type}")

Filling with pdfrw:

import pdfrw

ANNOT_KEY = '/Annots'
ANNOT_FIELD_KEY = '/T'
ANNOT_VAL_KEY = '/V'
ANNOT_RECT_KEY = '/Rect'
SUBTYPE_KEY = '/Subtype'
WIDGET_SUBTYPE_KEY = '/Widget'

template = pdfrw.PdfReader("sc100-blank.pdf")

data = {
    "(FieldName1)": "Value1",
    "(FieldName2)": "Value2",
}

for page in template.pages:
    annotations = page.get(ANNOT_KEY)
    if annotations:
        for annot in annotations:
            field_name = annot.get(ANNOT_FIELD_KEY)
            if field_name in data:
                annot.update(pdfrw.PdfDict(V=data[field_name]))
                # For text fields, also set appearance if needed
                annot.update(pdfrw.PdfDict(AP=''))  # Reset appearance

pdfrw.PdfWriter().write("sc100-filled.pdf", template)

Option 3: fillpdf / pdftk wrapper

from fillpdf import fillpdfs

# Get field names
fields = fillpdfs.get_form_fields("sc100-blank.pdf")
print(fields)

# Fill the form
data_dict = {
    "FieldName1": "Value1",
    "FieldName2": "Value2",
}
fillpdfs.write_fillable_pdf("sc100-blank.pdf", "sc100-filled.pdf", data_dict)

Option 4: pikepdf

import pikepdf

pdf = pikepdf.open("sc100-blank.pdf")

# Access AcroForm
if '/AcroForm' in pdf.Root:
    fields = pdf.Root.AcroForm.get('/Fields', [])
    for field_ref in fields:
        field = field_ref
        name = str(field.get('/T', ''))
        print(f"Field: {name}, Type: {field.get('/FT')}")

Handling Checkboxes

Checkboxes have specific "on" values. To find the valid value:

# With pypdf
fields = reader.get_fields()
for name, field in fields.items():
    if field.get('/FT') == '/Btn':
        # Check the appearance dictionary for valid states
        print(f"Checkbox: {name}")
        # The widget annotation's /AP/N dictionary keys show valid states

Common checkbox patterns:

Set to /Yes or the export value found in /AP/N keys
Uncheck by setting to /Off

For pypdf:

# To check a checkbox, you often need:
writer.update_page_form_field_values(
    writer.pages[0],
    {"CheckboxFieldName": "/Yes"},  # or "Yes" or True depending on form
    auto_regenerate=False
)

Key Tips

Always inspect field names first — PDF field names are not standardized.
Flatten vs non-flatten: Some approaches flatten the form (making fields non-editable). Use flatten=False if you want to keep fields editable.
Multi-page forms: Ensure you fill fields on the correct page.
Field hierarchy: Some PDFs use hierarchical field names like parent.child. The full qualified name matters.
Encoding: Use pdfrw.PdfString.encode() for pdfrw string values if needed.