> ## Documentation Index
> Fetch the complete documentation index at: https://docs.deepmerge.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Fetch one record by id

> What your AI tools can do with fetch.

<Note>Action name: `fetch` · Read only (never changes anything)</Note>

Fetch the full content of one memory page, or the transcript of one
cited source conversation, by id; ids come from `search` results and
from the \[\[id]] citations inside page bodies.
Returns { id, title, text, url, metadata }.

## Inputs

| Field | Type   | Required | What it is                      |
| ----- | ------ | -------- | ------------------------------- |
| `id`  | string | Yes      | Record id from a search result. |

<Accordion title="Full instructions the AI sees">
  ```text theme={null}
  This is also the ChatGPT deep-research / company-knowledge fetch step,
  with the result shape ChatGPT expects. Naming is fixed by the ChatGPT
  connector contract, so it deliberately breaks the get_ convention.

  USE WHEN: you have an id from a search result and need the full content.
  ```
</Accordion>