get_document: full text of one passage

Behaviour

Snippets in search results are the passage text, so most questions need no call here. Reach for it when you need the whole document rather than the matching passage.

Costs 1 credit.

Arguments

inputSchema

Name	Type	Description
`id`required	string	Passage id from a hit
`risk`	boolean	Also return risk_flags for the text: instruction_like, hidden_text, role_markers, low_trust_source
`strip`	boolean	Remove instruction-like sentences from the text (each becomes '[removed: instruction-like text]'). Implies risk.

Returns

A JSON payload in the text content block, and the same payload as structuredContent for clients that support it.

outputSchema

Name	Type	Description
`host`required	string	—
`host_rank`required	number	—
`id`required	string	—
`lang`required	string	—
`quality`required	integer	—
`risk_flags`	"instruction_like" \| "hidden_text" \| "role_markers" \| "low_trust_source"[]	Present when risk or strip was set. Computed on the original text.
`sentences_removed`	integer	Present when strip was set: how many sentences were removed
`text`required	string	The full cleaned body
`title`required	string	—
`url`required	string	—
`vertical`required	string	—

Calling it

Over Streamable HTTP, with the required arguments filled in:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_document",
    "arguments": {
      "id": "<id>"
    }
  }
}

Most clients do this for you — see Connect a client. To drive it by hand, complete the startup lifecycle first.

The same call over REST

If your client does not speak MCP, this capability is GET /doc/{id}, taking the same parameters and returning the same body.

curl -H "x-api-key: $UNLOB_API_KEY" \
  "https://api.unlob.com/doc/<id>"