Skip to content

Pdf Form Converter

github-actions[bot] edited this page Oct 29, 2026 · 3 revisions

PDF Form to Accessible HTML Converter

Overview

The convert_pdf_form_to_html MCP tool extracts AcroForm fields from a PDF and generates a fully accessible HTML5 form. It maps each PDF field type to semantically correct HTML elements with proper labels, fieldsets, ARIA attributes, and keyboard-friendly styling.

How It Works

  1. Reads the PDF file using pdf-lib (MIT license, pure JavaScript, no native dependencies)
  2. Extracts all AcroForm fields with their types, names, options, default values, and flags
  3. Derives human-readable labels from field names (splits camelCase, replaces underscores)
  4. Generates an accessible HTML5 form with proper structure

Field Type Mapping

Each PDF field type, with its pdf-lib class and HTML output.

PDF Field Type pdf-lib Class HTML Output
Text (single-line) PDFTextField <input type="text"> with <label>
Text (multiline) PDFTextField (multiline flag) <textarea> with <label>
Checkbox PDFCheckBox <input type="checkbox"> with <label>
Radio group PDFRadioGroup <fieldset> + <legend> + <input type="radio"> per option
Dropdown PDFDropdown <select> with <option> elements
Multi-select list PDFOptionList <select multiple> with <option> elements
Button PDFButton <button type="button">
Signature PDFSignature Placeholder paragraph with instructions

Accessibility Features in Generated HTML

  • Every input has an associated <label> with matching for/id binding
  • Radio button groups wrapped in <fieldset> with <legend>
  • Required fields marked with aria-required="true" and a visual asterisk indicator
  • Read-only fields use the readonly attribute with a visible "(read-only)" notice
  • Focus styles with 2px solid outline and offset for keyboard navigation
  • Input borders at 4.6:1 contrast ratio against white backgrounds
  • Responsive layout with max-width: 40rem and proper font sizing
  • Submit button meets 44x44px minimum touch target area

Usage

# Claude Code
convert_pdf_form_to_html filePath="forms/application.pdf" title="Job Application"

# Copilot Chat
/document-accessibility-wizard convert the enrollment PDF form to accessible HTML

Parameters

Each parameter, with its type, required and description.

Parameter Type Required Description
filePath string Yes Path to the PDF file containing form fields
title string No Title for the generated HTML page (default: filename)

Example Output

For a PDF with 3 text fields, 1 checkbox, and 1 dropdown, the tool generates:

PDF FORM CONVERSION: application.pdf
Fields extracted: 5 (text: 3, checkbox: 1, select: 1)

--- ACCESSIBLE HTML OUTPUT ---

<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <title>Job Application</title>
  ...
</head>
<body>
  <h1>Job Application</h1>
  <form action="#" method="post" novalidate>
    <div class="form-group">
      <label for="firstName">First Name</label>
      <input type="text" id="firstName" name="firstName">
    </div>
    ...
  </form>
</body>
</html>

--- FIELD INVENTORY ---

- firstName (text)
- lastName (text)
- email (text)
- agreeToTerms (checkbox)
- department (select) [5 options]

Limitations

Each form type, with its support level and notes.

Form Type Support Level Notes
Standard AcroForm Full Text, checkbox, radio, dropdown, multi-select, button
AcroForm with JavaScript actions Partial Fields extracted; calculated/validation JS requires manual review
XFA forms (XML Forms Architecture) Not supported Entirely different format; some government forms use this
Flattened forms Not supported Fields baked into content stream; no extractable data
Encrypted PDFs Attempted ignoreEncryption: true flag; may fail if fully locked

Security

  • Path validation: Files must be within the user's home directory or current working directory
  • Symlink resolution: Symlinks are followed and re-validated to prevent traversal
  • File size limit: 100 MB maximum to prevent memory exhaustion
  • HTML escaping: All field names, labels, and values are HTML-escaped to prevent XSS
  • No code execution: PDF JavaScript actions are not executed or converted

Installation

npm install pdf-lib

pdf-lib is a pure JavaScript library with zero native dependencies. It works on all platforms.

Connections

Each connection, with connect to and when.

Connect to When
pdf-accessibility Scan the source PDF for accessibility issues before conversion
forms-specialist Review the generated HTML form for additional accessibility improvements
web-issue-fixer Apply fixes to the generated HTML if issues are found

Clone this wiki locally