# PDFCheck - Full Technical Documentation

## Overview

PDFCheck (https://pdfcheck.online) is a privacy-focused PDF platform for analysis, validation, conversion, editing, and AI-assisted document review. Current account allowances and plan limits are documented on the pricing and API documentation pages.

## Core Tools

### 1. PDF Metadata Checker
**URL**: https://pdfcheck.online/

Extracts and analyzes standard and custom PDF metadata fields:
- **Creation & Modification Dates**: Shows the creation and modification dates stored in metadata and the time between them
- **Software Identification**: Detects specific software (Producer/Creator) used to generate the PDF
- **Author Information**: Extracts author names and contributor details
- **PDF Version**: Identifies PDF specification version (1.4, 1.7, 2.0)
- **Modification Status**: Flags when the modification date differs from the creation date
- **File Structure**: Analyzes linearization, page count, and layout

### 2. PDF Validator
**URL**: https://pdfcheck.online/pdf-validator

Runs basic PDF checks (not a formal conformance audit):
- **Parse Check**: Flags files our parser cannot open as corrupted or unreadable
- **File Type Check**: Confirms the file type is PDF
- **PDF/A Indicators**: Shows PDF version, tagging, and embedded fonts; use veraPDF for formal PDF/A conformance
- **Risky Content Check**: Flags JavaScript, actions, and embedded files

### 3. AI Content Detector
**URL**: https://pdfcheck.online/pdf-ai-detection

Checks PDF metadata for AI-generator indicators (indicators, not proof of AI writing):
- **Software Fingerprinting**: Detects ReportLab, PyPDF, WeasyPrint, pdf-lib, PDFKit, Puppeteer, and similar tools in the Creator and Producer fields
- **Metadata Heuristics**: Adds weight for a missing author or title and for automation keywords
- **Risk Score**: Provides a 0-100 risk score with the indicators found

### 4. AI PDF Chat
**URL**: https://pdfcheck.online/pdf-ai-chat

Chat with a PDF through PDFCheck-provided AI providers:
- **Plan Required**: Guests and free accounts can view the page; uploading and chatting require the AI plan
- **Monthly Allowance**: The AI plan includes 1,000 successful AI messages per month and unlimited one-PDF chat sessions
- **Local Retrieval**: PDFCheck extracts page text into private source chunks and sends only relevant excerpts to OpenAI or DeepSeek
- **Grounded Answers**: Answers include validated page citations from the uploaded PDF
- **PDF Review Workspace**: Shows the PDF preview beside chat, extracted source chunks, selected context, comments, and AI-generated document notes
- **OCR Fallback**: If normal extraction finds no text, PDFCheck can render PDF pages as JPG images and send them to the configured OpenAI OCR model
- **Markdown Export**: Users can download extracted content and annotations as Markdown
- **Review Prompts**: Summaries, important sections, grammar issues, internal consistency, key dates/numbers, risks, and simple explanations
- **Retention**: Original PDFs are stored privately until the user deletes the chat

### 5. PDF Optimizer
**URL**: https://pdfcheck.online/optimize-pdf

Compresses and reduces PDF file size:
- **Screen (72dpi)**: Maximum compression for web viewing
- **eBook (150dpi)**: Balanced quality and size
- **Printer (300dpi)**: Minimal compression for printing
- **Prepress (300dpi)**: High quality for professional printing

### 6. PDF Signature Checker
**URL**: https://pdfcheck.online/pdf-signature-checker

Inspects digital signatures and certificates in PDF documents (no cryptographic or trust-chain validation):
- **Signature Detection**: Identifies digital signatures in a PDF
- **Certificate Details**: Extracts signer name, organization, validity dates, and algorithm
- **Byte Range Coverage**: Checks whether the signed byte range reaches the end of the file or bytes were added after signing
- **Timestamp Token Detection**: Shows whether a timestamp token is present and whether the certificate dates include today

### 7. PDF Image Extractor
**URL**: https://pdfcheck.online/extract-images

Extracts all images from PDF documents:
- **Full Extraction**: Extracts every embedded image from all pages
- **Original Quality**: Preserves original image resolution and format
- **Batch Download**: Download all extracted images as a ZIP archive
- **Format Support**: Handles JPEG, PNG, TIFF, and other embedded image formats
- **Preview Gallery**: Visual preview of all extracted images before download

### 8. PDF Accessibility Checker
**URL**: https://pdfcheck.online/pdf-accessibility-checker

Checks basic PDF accessibility signals (not a PDF/UA or WCAG audit):
- **Tagged PDF Detection**: Checks whether the PDF is tagged (tag quality and reading order are not reviewed)
- **Document Title**: Checks for a meaningful document title in metadata
- **Language Declaration**: Validates that the document language is specified
- **Text Content**: Confirms the PDF contains extractable text (not image-only)
- **Bookmarks/Navigation**: Checks for document outline and navigation aids
- **Accessibility Score**: Provides an overall score (0-5) with actionable recommendations

### 9. PDF Form Filler
**URL**: https://pdfcheck.online/fill-pdf-form

Fill in interactive PDF form fields online. Upload a PDF with form fields, fill them in via the browser, and download the completed document.

### 10. PDF Form Builder
**URL**: https://pdfcheck.online/build-pdf-form

Add fillable form fields (text inputs, checkboxes, dropdowns) to any PDF document. Upload a PDF, visually place form fields, and export as an interactive fillable PDF.

### 11. PDF Editor
**URL**: https://pdfcheck.online/edit-pdf

Edit PDF text, images, and annotations online. Upload a PDF, make edits directly in the browser, and download the modified document.

### 12. PDF Highlighter
**URL**: https://pdfcheck.online/highlight-pdf

Highlight and annotate PDF files online. Upload a PDF, draw transparent highlights, add text notes or rectangle callouts, save versions, and download the annotated PDF.

### Additional PDF Workflows

- **PDF/A Validator**: https://pdfcheck.online/pdf-a-validator
- **Batch PDF Checker**: https://pdfcheck.online/batch-pdf-checker
- **PDF Text Extractor**: https://pdfcheck.online/pdf-text-extract
- **PDF to Markdown**: https://pdfcheck.online/pdf-to-markdown
- **PDF Font Checker**: https://pdfcheck.online/pdf-font-checker
- **Compare PDFs**: https://pdfcheck.online/pdf-compare
- **PDF to Image**: https://pdfcheck.online/pdf-to-image
- **Merge PDF**: https://pdfcheck.online/merge-pdf
- **Split PDF**: https://pdfcheck.online/split-pdf
- **PDF Password Tool**: https://pdfcheck.online/pdf-password
- **PDF Metadata Editor**: https://pdfcheck.online/pdf-metadata-editor
- **PDF Metadata Remover**: https://pdfcheck.online/pdf-metadata-remover
- **PDF Converter Directory**: https://pdfcheck.online/pdf-converter
- **All PDF Tools**: https://pdfcheck.online/all-pdf-tools

## REST API

### Authentication
Anonymous metadata analysis is available without an account. Register at https://pdfcheck.online/register and generate a token at https://pdfcheck.online/account/api-tokens for account access. Include it in the Authorization header:
```
Authorization: Bearer YOUR_API_TOKEN
```

### Endpoints

#### Analyze PDF
```
POST /api/v1/pdf/analyze
Content-Type: multipart/form-data

Parameters:
- file: PDF file (required; Free/anonymous: up to 50 MB, new Pro/AI: up to 100 MB)

Response: JSON with success, data.id, data.filename, data.file_size,
data.metadata.basic, data.metadata.document, data.metadata.dates,
data.analyzed_at, and remaining_checks
```

Read page count from data.metadata.basic.pages, PDF version from data.metadata.basic.pdf_version, and author from data.metadata.document.author. Date objects include iso8601 and parsed fields. The API creates private analyses; it does not accept an is_public parameter. Metadata is editable and does not prove document authenticity or AI authorship.

#### Get Analysis Results
```
GET /api/v1/pdf/{token}
Authorization: Bearer YOUR_API_TOKEN

Response: JSON with full analysis data
```

Use the returned data.id as token. Authentication is required, and the result must belong to your account or be public. Use your account token when analyzing to enable later retrieval. Newly created anonymous results are private and cannot be retrieved later; retain their response in your client.

#### List Analyses
```
GET /api/v1/pdf
Authorization: Bearer YOUR_API_TOKEN

Response: JSON with data entries and pagination metadata
```

#### Get User Info
```
GET /api/v1/user
Authorization: Bearer YOUR_API_TOKEN

Response: JSON with usage stats, remaining quota
```

#### Extract Text or Markdown
```
POST /api/v1/pdf/extract-text
POST /api/v1/pdf/to-markdown
Content-Type: multipart/form-data

Parameters:
- pdf: PDF file (required)
```

These endpoints use pdf, not the metadata endpoint's file field. Each returns success and data; Markdown also returns remaining_conversions. Handle unsuccessful responses and tool-specific limits before using the result.

### Rate Limits
Anonymous metadata analysis allows 10 checks per day per IP across MCP and REST. Free account tokens have 100 metadata checks per day; paid accounts use their plan allowances. Markdown has a separate allowance, including 10 conversions per day per IP anonymously. Other limits vary by tool. Check the response and account usage rather than assuming one quota applies to every operation.

## Privacy & Security

- **File Handling**: Request-only checker uploads are not retained after processing; editors, converters, OCR, and other workflows can use temporary or persistent storage as disclosed in the Privacy Policy
- **HTTPS**: All data transferred via encrypted connection
- **Data Retention**: Temporary files have a 24-hour cleanup ceiling; guest metadata/comparison results are retained for 90 days; account and AI Chat data follow their documented deletion controls
- **Shareable Links**: Results can be shared via unique tokens

## Languages

PDFCheck is available in 11 languages. Each language has its own homepage:

- [English](https://pdfcheck.online/)
- [Hindi (हिन्दी)](https://pdfcheck.online/hi)
- [Danish (Dansk)](https://pdfcheck.online/da)
- [German (Deutsch)](https://pdfcheck.online/de)
- [Spanish (Español)](https://pdfcheck.online/es)
- [French (Français)](https://pdfcheck.online/fr)
- [Italian (Italiano)](https://pdfcheck.online/it)
- [Japanese (日本語)](https://pdfcheck.online/ja)
- [Korean (한국어)](https://pdfcheck.online/ko)
- [Lithuanian (Lietuvių)](https://pdfcheck.online/lt)
- [Chinese, Simplified (中文)](https://pdfcheck.online/zh)

Tool pages are available under the same language prefix, for example https://pdfcheck.online/hi/merge-pdf or https://pdfcheck.online/de/pdf-validator. Guides, blog articles and methodology pages are in English.

## Resources

- **How It Works**: https://pdfcheck.online/how-it-works
- **API Documentation**: https://pdfcheck.online/api-docs
- **MCP Server**: https://pdfcheck.online/mcp-server
- **AI Agents Guide**: https://pdfcheck.online/ai-agents
- **Sitemap**: https://pdfcheck.online/sitemap
- **About**: https://pdfcheck.online/about
- **Privacy Policy**: https://pdfcheck.online/privacy

## AI Agent Integration

### Remote MCP Server
Endpoint: https://pdfcheck.online/api/mcp (HTTP; no local server script required).
Use a client that supports remote HTTP MCP and optional bearer headers. Configuration formats vary by client; this setup does not support OAuth-only connections.

Example configuration for a client using mcpServers:
```json
{
  "mcpServers": {
    "pdfcheck": {
      "url": "https://pdfcheck.online/api/mcp"
    }
  }
}
```

For account access, add a headers object under pdfcheck with Authorization set to Bearer YOUR_API_TOKEN. Keep tokens in your client's supported secret settings where available.

Available PDF MCP tools:
- analyze_pdf: requires file_base64 and filename. The client must read the PDF and base64-encode its contents; a local file path alone will not work.
- get_analysis: takes token, using data.id from the analysis response. Authenticated users can retrieve their own or public analyses; anonymous users can retrieve public analyses only.
- list_analyses: requires an account token; returns up to 50 recent account analyses.
- check_usage: returns account or anonymous metadata allowance.

Read the JSON tool result from content[0].text. Use an account token when analyzing for later account retrieval. New anonymous results are private and must be retained by the client.

MCP does not expose text extraction, Markdown conversion, compression, merging, signing, or formal validation. Browser form URLs are not public REST endpoints. Metadata indicators cannot establish AI authorship, authenticity, signature trust, PDF/A or PDF/UA compliance, or accessibility.

The optional local stdio bridge uses a different analyze_pdf argument, filePath, and uploads the file to REST. Its history and usage methods require an account token.

### REST API
Base URL: https://pdfcheck.online/api/v1
- POST /pdf/analyze — Inspect metadata (multipart field file)
- POST /pdf/extract-text — Extract text (multipart field pdf)
- POST /pdf/to-markdown — Convert to Markdown (multipart field pdf)
- GET /pdf/{token} — Retrieve an account-owned or public analysis; bearer token required
- GET /pdf — List account analyses; bearer token required
- GET /user — Check account quota; bearer token required

### Shell Wrapper
This is a small shell function, not a separate CLI package. It requires cURL. Set PDFCHECK_API_TOKEN in your environment and call pdfcheck ./document.pdf:
```bash
pdfcheck() {
  if [ "$#" -ne 1 ] || [ ! -f "$1" ]; then
    printf '%s\n' 'Usage: pdfcheck /path/to/document.pdf' >&2
    return 2
  fi
  if [ -z "${PDFCHECK_API_TOKEN:-}" ]; then
    printf '%s\n' 'Set PDFCHECK_API_TOKEN before calling pdfcheck.' >&2
    return 2
  fi
  curl --fail-with-body --silent --show-error \
    "https://pdfcheck.online/api/v1/pdf/analyze" \
    -H "Authorization: Bearer $PDFCHECK_API_TOKEN" \
    -H "Accept: application/json" \
    -F "file=@$1"
}
```

### Rate Limits
- API tokens: Generate at https://pdfcheck.online/account/api-tokens

### Machine-Readable Discovery
- LLM summary: https://pdfcheck.online/llms.txt
- XML sitemap index: https://pdfcheck.online/sitemap.xml
- API catalog: https://pdfcheck.online/.well-known/api-catalog
- Agent skills: https://pdfcheck.online/.well-known/agent-skills/index.json
- MCP server card: https://pdfcheck.online/.well-known/mcp/server-card.json
- OAuth protected-resource metadata: https://pdfcheck.online/.well-known/oauth-protected-resource

### PDF Skills for AI Agents
Copyable task instructions are available at https://pdfcheck.online/ai-agents#pdf-skills. A prompt does not install an integration or grant file access:
- PDF Analysis Skill — Inspect metadata with MCP or REST and report missing fields and limitations
- PDF Generate Skill — Generate and visually review a PDF using your own tools, then inspect its metadata with PDFCheck
- PDF Review Skill — Use PDFCheck for metadata and separate specialist tools for structure, accessibility, or formal compliance
- PDF Workflow Skill — Combine REST metadata/text extraction with your own routing, storage, and review tools

## Technical Details

### Supported PDF Versions
PDF 1.0 through PDF 2.0

### File Size Limits
- Web upload: 50MB on the free plan, 100MB on Pro and AI
- API upload: 50MB on the free plan, 100MB on Pro and AI

### Analysis Methods
1. XMP metadata extraction
2. Document Information Dictionary parsing
3. Software signature matching
4. Creation and modification date comparison
5. Risky content marker scan (JavaScript, actions, embedded files)

## Contact

- **Website**: https://pdfcheck.online
- **Email**: support@businesspress.io