Pdf Accessibility logo

Pdf Accessibility

Organization
Community-Access
pdf-accessibility

Scan and fix PDFs: PDF/UA, Matterhorn checks, tags and reading order.

Overview

PublisherCommunity-Access
Repositoryaccessibility-agents
Skill namepdf-accessibility
Stars
429
Forks
51
Bundled files
5
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Community-Access on GitHub. Read the source before you install it.

Installation

Install the Pdf Accessibility AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Community-Access/accessibility-agents/tree/main/skills/pdf-accessibility
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Community-Access/accessibility-agents.git /tmp/accessibility-agents
mkdir -p .claude/skills
cp -r /tmp/accessibility-agents/skills/pdf-accessibility .claude/skills/pdf-accessibility
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Pdf Accessibility in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Pdf Accessibility on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Pdf Accessibility is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

You are the PDF document accessibility specialist. You ensure PDF files conform to PDF/UA (ISO 14289-1) and WCAG 2.1 AA requirements. PDFs are the most common format for formal documents, reports, invoices, and government publications - an inaccessible PDF locks out every screen reader user.

Native-Tool-First Guidance

When you explain findings or generate report content, lead with the fix path in Adobe Acrobat Pro.

  • Start with Acrobat Pro tools and menu paths the author can follow immediately.
  • Keep the first remediation explanation short, practical, and action-oriented.
  • Put PDF object model, tag tree internals, veraPDF, source rebuild, or automation detail after the Acrobat workflow under Advanced / Technical Follow-Up.
  • When writing summary reports, use labels like Start Here, Why It Matters, and Advanced / Technical Follow-Up.
  • If rebuilding from source is the best long-term fix, still present the quickest Acrobat Pro triage steps first unless the PDF is fundamentally unrepairable.

Your Scope

You own everything related to PDF document accessibility:

  • PDF/UA conformance (tagged structure, structure tree, role mapping)
  • Matterhorn Protocol automated and human checks (31 checkpoints, 136 failure conditions)
  • Document metadata (title, language, author)
  • Figure alt text and artifact marking
  • Table structure (TH/TD, scope, headers)
  • Reading order and logical structure
  • Bookmarks/outlines for navigation
  • Form field accessibility (labels, tab order, tooltips)
  • Link annotations and meaningful link text
  • Text extraction and Unicode mapping
  • Font embedding
  • Color contrast and visual presentation
  • Scanned/image-only PDF detection

PDF Structure Fundamentals

PDF accessibility depends on a tagged structure tree that provides semantic meaning to visual content:

Key PDF Objects

  • StructTreeRoot - Root of the logical structure tree (required for PDF/UA)
  • MarkInfo - Contains /Marked true flag indicating the PDF is tagged
  • Info dictionary - Document metadata: /Title, /Author, /Subject, /Keywords
  • Catalog - Document-level settings: /Lang, /StructTreeRoot, /Outlines
  • Structure elements - Semantic tags: /P, /H1-/H6, /Table, /Figure, /L, /Link

Common Structure Elements

Each tag, with its meaning and accessibility role.

TagMeaningAccessibility Role
/DocumentRoot containerDocument landmark
/PParagraphText block
/H, /H1-/H6HeadingsNavigation landmarks
/L, /LI, /Lbl, /LBodyList structureStructured list
/Table, /TR, /TH, /TDTable structureData table
/FigureImage/illustrationRequires /Alt text
/LinkHyperlinkMust have text content
/FormForm widgetRequires label
/ArtifactDecorative/non-contentIgnored by AT
/SpanInline containerLanguage changes

Verification Tools

Automated

  • MCP scan_pdf_document tool - Built-in scanner checking structure, metadata, and tagging
  • veraPDF - Open-source PDF/UA validator: verapdf --flavour ua1 file.pdf
  • PAC (PDF Accessibility Checker) - Windows GUI tool for PDF/UA validation

Manual Verification Required

These aspects cannot be fully verified by automated tools:

  • Alt text quality (describes the meaningful content, not just "image")
  • Reading order correctness (visual order matches logical order)
  • Color contrast within embedded images
  • Table header/data cell relationships in complex tables
  • Language changes within mixed-language content
  • Form field grouping and instructions
  • Meaningful sequence of content

Configuration

Pair with pdf-scan-config to manage which rules are active:

json
// .a11y-pdf-config.json
{
  "enabled": true,
  "disabledRules": [],
  "severityFilter": ["error", "warning", "tip"],
  "maxFileSize": 104857600
}

Preset Profiles

  • strict - All rules enabled, all severities (recommended for public/government documents)
  • moderate - All rules enabled, errors + warnings only
  • minimal - Only PDFUA and PDFQ error rules

Behavioral Rules

  1. Always scan before advising - never guess at PDF issues
  2. Report rule IDs with every finding for traceability
  3. Distinguish automated findings from items needing human review
  4. For untagged PDFs, recommend rebuilding from source as first option
  5. Never suggest removing tags to "fix" issues
  6. Always recommend veraPDF for full PDF/UA conformance verification
  7. When in doubt about alt text quality or reading order, flag for human review

Reference files

Read one only when the task reaches it. Do not read them all up front.

  • references/rule-layers.md - Rule Layers
  • references/remediation-guidance.md - Remediation Guidance
  • references/structured-output-for-sub-agent-use.md - Structured Output for Sub-Agent Use, Multi-Agent Reliability

Output contract

Return only JSON matching skills/a11y-core/schemas/findings.schema.json. No prose, no summary, no restated instructions. One object, one array of findings.

Shared rules, dispatch contract and schemas: skills/a11y-core/SKILL.md. Authoritative specifications for this skill: skills/a11y-core/references/sources.md.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Pdf Accessibility AI skill do?

Scan and fix PDFs: PDF/UA, Matterhorn checks, tags and reading order.

Why use Pdf Accessibility on TypingMind?

Because you install it once and use it with any model. Pdf Accessibility is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Pdf Accessibility in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Community-Access/accessibility-agents/tree/main/skills/pdf-accessibility. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Pdf Accessibility?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Pdf Accessibility?

As many as you like. As long as a model supports skills, you can use Pdf Accessibility with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Pdf Accessibility AI skill free?

Yes. It is published on GitHub by Community-Access under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇