Extract Text from PDFs
Pull raw text from any PDF URL so you can search, quote, or process its contents programmatically.
Get PDF Metadata and Info
Retrieve metadata from any PDF URL including page count, author, creation date, and file size.
Tool
Before processing a PDF — extracting its text, merging it, or sending it to a client — you often need basic facts: how many pages is it, when was it created, who authored it, and is it a scanned image or contains real text? Downloading the file just to check these details is wasteful.
The get_info skill fetches metadata from any public PDF URL without downloading the full file. You get page count, file size, creation and modification dates, author, title, and producer information immediately. This lets you validate documents before processing them and surface basic attribution for cataloguing.
Document management teams validating submissions, developers building PDF processing pipelines, and researchers cataloguing document collections use this to triage files before committing to heavier processing.
Agent Guides
Claude
- Connect ToolRouter to Claude: claude mcp add toolrouter -- npx -y toolrouter-mcp
- Ask: "Get info about this PDF: [paste URL]"
- Claude returns metadata including page count, author, and creation date
ChatGPT
- Add ToolRouter to ChatGPT using the MCP JSON configuration
- Ask: "What are the properties of this PDF: [url]?"
- ChatGPT retrieves and displays the metadata
Copilot
- Add ToolRouter to your Copilot MCP configuration
- In Copilot Chat: "Get the metadata for this PDF: [url]"
- Copilot returns the document properties
OpenClaw
- Connect ToolRouter to OpenClaw: openclaw mcp add toolrouter -- npx -y toolrouter-mcp
- Provide a list of PDF URLs: "Get metadata for each of these documents"
- OpenClaw retrieves properties for all files
Related Use Cases
Summarise PDF Documents
Get AI-generated summaries of long PDF reports, contracts, or research papers from any public URL.
Merge Multiple PDFs
Combine several PDF files from public URLs into a single merged document.