Integration

PDF Vector integration

PDF Vector turns documents into text and data with AI: parse PDFs, Office files, images and more to markdown, ask questions about a document, or extract structured JSON matching your own JSON Schema, with invoice-tuned variants. It also searches academic papers (Semantic Scholar, PubMed, arXiv, OpenAlex and more), fetches papers by DOI or id, builds citation graphs, finds similar papers and citations for a paragraph, parses papers to markdown and searches research grants. Every call spends PDF Vector credits. Documents are passed as a public URL or base64 (no multipart upload); the identity-document and bank-statement routes are not exposed, nor the async callback mode. Authentication is a PDF Vector API key sent as a Bearer token.

Connect PDF Vector How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in PDF Vector.Built from the 13 PDF Vector tools below, on the connection you add once.

What you can do with PDF Vector

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Paper, 5 operations

fetch_papersFetch papersPOSTFetches up to 100 papers by identifier (DOI, PubMed, arXiv, Semantic Scholar, ERIC, Europe PMC or OpenAlex id), detecting the database itself. 2 credits per request.
find_similar_papersFind similar papersPOSTFinds papers similar to a seed paper through its citation network, ranked by similarity, optionally with citation edges for a graph. 3 credits per request.
get_paper_graphGet paper graphPOSTReturns a paper with the papers citing it and the papers it references, paged by offset. Spends max(2, ceil(returned citations plus references / 50)) credits per request, so 2 credits at least.
parse_paperParse paperPOSTFinds the public PDF of a paper from its identifier or URL, parses it and returns the paper as markdown. Fails with NOT_FOUND when no public PDF exists. Billed per page by model tier.
search_papersSearch papersPOSTSearches academic papers across databases (Semantic Scholar by default, PubMed, arXiv, OpenAlex and more) with relevance ranking and year filters. 2 credits per request.

Document, 3 operations

ask_documentAsk documentPOSTAnswers a question about a document with AI and returns the answer as markdown. Pass the document as url (public link) or base64. Billed per page of the document by model tier (nano cheapest).
extract_documentExtract documentPOSTExtracts structured data from a document into JSON matching the JSON Schema you give (schema) and an instruction (prompt). Result in data. Pass the document as url (public link) or base64. Billed per page by model tier (nano cheapest, 2 credits).
parse_documentParse documentPOSTParses a document to markdown and returns the text, page count and credits used. Pass the document as url (public link) or base64. Billed per page by model tier (nano cheapest).

Invoice, 3 operations

ask_invoiceAsk invoicePOSTAnswers a question about an invoice with AI and returns the answer as markdown. Pass the document as url (public link) or base64. Billed per page by model tier (nano cheapest, 6 credits).
extract_invoiceExtract invoicePOSTExtracts structured data from an invoice into JSON matching the JSON Schema you give (schema) and an instruction (prompt). Result in data. Pass the document as url (public link) or base64. Billed per page by model tier (nano cheapest, 6 credits).
parse_invoiceParse invoicePOSTParses an invoice to markdown with invoice-tuned extraction (pro or max tier). Pass the document as url (public link) or base64. Billed per page: pro 6 credits, max 10.

Citation, 1 operation

find_citationsFind citationsPOSTFinds papers to cite for a paragraph: splits it into sentences and returns scored citations (0 to 10) for each. Spends 2 credits per sentence for which citations are found; sentences with none are free.

Grant, 1 operation

search_grantsSearch grantsPOSTSearches research grants and funding calls in Grants.gov, NIH RePORTER, CORDIS and UKRI, filterable by deadline and funding amount. 2 credits per request.

Use PDF Vector from Claude Code, Cursor or Codex

Connect PDF Vector in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its PDF Vector operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect PDF Vector

Create your workspace
Sign up free. Nothing to install.
Connect PDF Vector once
Add your PDF Vector connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.