PDF — Read, analyze & merge PDFs

Tools / PDF

Extract Text

Pull the text out of a PDF so you can search, quote, or process the content directly.

Returns: The extracted text along with page count and file size

Parameters

url *stringURL of the PDF to extract text from. The tool server CAN fetch any public URL — always pass it.

pagesstringPage range to extract (e.g. "1-5", "1,3,7-9"). Omit for all pages.

Example

Get the text from a research paper

curl -H "Authorization: Bearer $TOOLROUTER_API_KEY" \
  -d '{
  "tool": "pdf",
  "skill": "extract_text",
  "input": {
    "url": "https://example.com/paper.pdf"
  }
}' \
  https://api.toolrouter.com/v1/tools/call

Analyze PDF

Ask a question about a PDF and get an AI-written answer — summaries, data extraction, contract review, or any analysis you describe in your prompt.

Returns: AI-written analysis in markdown with token usage details

Parameters

url *stringURL of the PDF to analyze. The tool server CAN fetch any public URL — always pass it.

promptstringAnalysis instruction — what you want to know about the document. Defaults to a thorough summary.

pagesstringPage range to analyze (e.g. "1-5", "3,7-10"). Omit for all pages.

modelstringAI model override in provider/model format (e.g. "anthropic/claude-3.5-sonnet"). Uses a fast, capable default.

Example

Give me a summary of this research paper

curl -H "Authorization: Bearer $TOOLROUTER_API_KEY" \
  -d '{
  "tool": "pdf",
  "skill": "analyze",
  "input": {
    "url": "https://example.com/paper.pdf",
    "prompt": "Summarize the key findings and methodology"
  }
}' \
  https://api.toolrouter.com/v1/tools/call

Get PDF Info

Check a PDF for page count, title, author, file size, and creation date without reading the full content.

Returns: Page count, author, title, file size, and other document properties

Parameters

url *stringURL of the PDF to inspect. The tool server CAN fetch any public URL — always pass it.

Example

How long is this PDF and who wrote it?

curl -H "Authorization: Bearer $TOOLROUTER_API_KEY" \
  -d '{
  "tool": "pdf",
  "skill": "get_info",
  "input": {
    "url": "https://example.com/document.pdf"
  }
}' \
  https://api.toolrouter.com/v1/tools/call

Merge PDFs

Join multiple PDFs into one downloadable file so you can bundle reports, combine chapters, or consolidate documents.

Returns: A single merged PDF with total page count and file size

Parameters

urls *arrayList of PDF URLs to merge (2-20). The tool server CAN fetch any public URL — always pass them.

Example

Combine the Q1 and Q2 reports into a single PDF

curl -H "Authorization: Bearer $TOOLROUTER_API_KEY" \
  -d '{
  "tool": "pdf",
  "skill": "merge",
  "input": {
    "urls": [
      "https://example.com/report-q1.pdf",
      "https://example.com/report-q2.pdf"
    ]
  }
}' \
  https://api.toolrouter.com/v1/tools/call

Quick Start

MCP (Claude Code)

claude mcp add --transport stdio \
  --env TOOLROUTER_API_KEY=YOUR_API_KEY \
  toolrouter -- npx -y toolrouter-mcp

REST API

curl -H "Authorization: Bearer $TOOLROUTER_API_KEY" \
  -d '{"tool":"pdf","skill":"extract_text","input":{}}' \
  https://api.toolrouter.com/v1/tools/call

Frequently Asked Questions

Can I check a PDF without reading the whole file?

Yes. `get_info` returns metadata such as page count, title, author, file size, and creation date without extracting the full content.

How do I pull text from only part of a PDF?

Use `extract_text` with a `pages` range such as `1-10` or `3,7,12-15` when you only need a section of the document.

Can I ask a specific question about a PDF?

Yes. `analyze` accepts a prompt like 'extract the revenue figures into a table' or 'summarize the key findings' and returns an AI-written answer.

Can I merge several PDFs into one file?

Yes. `merge` combines 2 to 20 public PDF URLs into a single downloadable file and returns the merged page count and file size.