What it does
PDF Kit for Agents does four jobs on PDF files that sit at a public web address. Info: how many pages, page sizes, version and the document's own details. Text: the words on the pages you choose. Merge: put two to ten PDFs together in order, each with the pages you want. Split: keep just the pages you choose, or one file per page. Every answer carries a signed, dated receipt with the SHA-256 and size of each file that went in and each file or text that came out, so either side can later show exactly what was processed. Same input and options give the same output file. No model, no sign-up, nothing stored.
- Who it is for
- Agents and people who handle stacks of PDFs (filings, reports, contracts, statements) and need the page count, the text, a combined file or a few pages, with a checkable record of what went in and what came out.
- Price
- Free, no sign-up: 50 jobs a day per caller, up to 20 MB per file. Paid calls are not offered.
- Limits
- Public web addresses only. Up to 20 MB per file, 40 MB in all, 1,000 pages per file; a PDF up to 10 MB comes back in the answer, bigger results are refused, never cut. Encrypted PDFs are refused. No OCR: a scanned page is a picture, so text returns nothing for it and the answer says so; text in fonts that carry no Unicode map is missed, and content in the LZW compression format is missed (the answer warns). Info on several addresses counts one job per address. Text order follows how the file draws it, which for columns and tables can differ from reading order. Merge and split copy each page's drawn content and resources unchanged into a fresh PDF; ALL annotations (links, comments, stamps, highlights, signatures, form fields), bookmarks, tags and the source's document details are NOT carried over. The receipt proves we processed files with those hashes at that time and produced outputs with those hashes. It does not prove what the files say or who made them. At most 120 fetches an hour from one web server.
For developers: try it
POST https://llpro.ai/mcp, tools/call pdf_kit with action (info, text, merge, split) and url, or urls for merge (each an address or {url, pages}); optional pages like '1-3,5' and each=true for split. -> pages / text / pdf_base64, the SHA-256 of each file, signed observation, receipt, stamp link.For agents
Tool pdf_kit on https://llpro.ai/mcp and in the tools list. Signed answers are checked with the public key below (signer llpro1:kvh4snd6pgqrc4itkjenzhd6e4): an Ed25519 signature over the observation as canonical JSON (sorted keys, no spaces).
Public key (base64): cAKC75+9whNsCUKlD3zWJDlHHoc2AyYFgeUpXQhwcts=
This tool exists to be found by agents that search MCP registries for a PDF merge, split, text-extraction or page-count tool (road: MCP registries, through the Campus gateway: MCP, Agent2Agent and REST). One free call, one signed answer with the SHA-256 of every input and output. Keep the signed observation with the files. How a stamp is checked: https://llpro.ai/stamp
Next
- Connect your agent: Use this and the other LLPRO tools from your own agent.
- Get your agent: No agent yet? Get one that asks first.
- Hand the whole job to the Hub: A PDF is one step. Tell the Hub the job (assemble a filing, read a stack of reports, answer questions from contracts); it routes you to the right tools.
For agents: this page as Markdown · JSON · llms.txt · MCP endpoint llpro.ai/mcp · agent card