Integration

Typless integration

Typless extracts structured data from business documents (invoices, receipts, delivery notes, bank statements, contracts, purchase orders and other custom documents). Run a pretrained model on a document by public URL, or extract with one of your own document types (fields, line items and VAT rates defined in the Typless app), fetch finished asynchronous results, send corrected values back as feedback, add documents to a document type's dataset and start training. Creating and editing document types, webhooks and plugins is done in the Typless app and is not exposed. The extraction and add-document routes take the file as base64 text, which the platform cannot produce from a stored file; for a stored file, make it reachable by a public URL and use the pretrained-model route, which takes file_url. Authentication is an API key from the Typless app (Settings, API keys) sent in the Authorization header as 'Token <key>'.

Connect Typless How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Typless.Built from the 9 Typless tools below, on the connection you add once.

What you can do with Typless

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.

Document extraction, 3 operations

extract_document_dataExtract document dataPOSTExtracts data from one document with one of your document types and returns the fields, line items and VAT rates found, each value with its confidence. The file goes in as base64 text.
extract_document_data_asyncExtract document data asyncPOSTSame extraction as extract_document_data (file as base64 text), run as a background job on Typless; can also return the document's text blocks. Waits for the job and returns the finished result; gives up after 10 minutes.
extract_with_pretrained_modelExtract with pretrained modelPOSTRuns a Typless pretrained model (invoice, receipt, delivery_note, packing_list, bank_statement, transport_report, contract, purchase_order) on file_url or a base64 file. Waits for the job and returns the finished result; gives up after 10 minutes.

Training document, 2 operations

add_training_documentAdd training documentPOSTAdds a document with its correct field values to a document type's dataset (training data), for example to set up a new document type from documents already processed. The file goes in as base64 text.
add_training_document_asyncAdd training document asyncPOSTSame as add_training_document but Typless processes the document in the background; the answer only confirms acceptance. The file goes in as base64 text.

Extraction, 1 operation

list_extractions_awaiting_pollList extractions awaiting pollGETLists the extraction ids of asynchronous extractions that have finished but whose result has not been fetched yet. Fetch each with get_extraction_result.

Extraction feedback, 1 operation

add_extraction_feedbackAdd extraction feedbackPOSTSends the correct values for a document extracted earlier (by its object_id), adding it to the document type's dataset so the model learns from it. Send every field of the document type.

Extraction result, 1 operation

get_extraction_resultGet extraction resultGETReturns an asynchronous extraction by extraction_id: status IN_PROGRESS, SUCCESS (result filled), ERROR (error filled) or EXPIRED (deleted 48 hours after it finished). Use it for ids from list_extractions_awaiting_poll.

Training, 1 operation

start_trainingStart trainingPOSTStarts training of every model of the named document type on its current dataset. The answer lists the supplier models whose training started.

Use Typless from Claude Code, Cursor or Codex

Connect Typless in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Typless 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 Typless

Create your workspace
Sign up free. Nothing to install.
Connect Typless once
Add your Typless 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.