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

# bigdata_upload_document

> Upload and index UTF-8 text into the private Bigdata.com corpus.

## Overview

Upload Document saves content into the user's private [Bigdata.com](https://bigdata.com) corpus and submits it for enrichment and indexing. It is designed for research artifacts — memos, summaries, screens, comparisons, and analyses — so they can be retrieved later with [`bigdata_search`](/mcp-reference/tools/bigdata-search) and the private content discovery tools.

## When to Use

Upload Document is ideal for:

* **Saving research artifacts:** Persisting memos, summaries, or analyses produced during a session
* **Building a private corpus:** Adding tagged notes that remain searchable across future workflows
* **Sharing within the organisation:** Optionally making uploaded content available to other members of the organisation

<Tip>
  This tool accepts **UTF-8 text only** (Markdown preferred, HTML, or plain text). For original binary files of any type or size, use [https://app.bigdata.com/files](https://app.bigdata.com/files). You can also extract text from a PDF or similar file and upload that text as Markdown.
</Tip>

## How It Works

1. **Prepare the content** as UTF-8 text (Markdown preferred)
2. **Choose tags** using the `prefix:value` convention (for example `topic:semiconductors`, `project:q3-review`). Optionally review existing tags with [`bigdata_list_tags`](/mcp-reference/tools/bigdata-list-tags)
3. **Decide sharing:** leave private by default, or share with the organisation
4. **Upload** with `bigdata_upload_document` — the response returns immediately while enrichment and indexing continue in the background
5. **Check status** later with [`bigdata_list_documents`](/mcp-reference/tools/bigdata-list-documents), or open the file at [https://app.bigdata.com/files](https://app.bigdata.com/files)

## Parameters

| Parameter        | Type      | Required | Description                                                                                                     |
| ---------------- | --------- | -------- | --------------------------------------------------------------------------------------------------------------- |
| `content`        | string    | Yes      | Document body as UTF-8 text: Markdown (preferred), HTML, or plain text.                                         |
| `file_name`      | string    | Yes      | Display name inside Bigdata.com (for example `NVDA Earnings Preview Q2 2025`).                                  |
| `tags`           | string\[] | No       | Tags to apply. Prefer `prefix:value` so they are easy to filter later. Only the tags provided here are applied. |
| `share_with_org` | boolean   | No       | If `true`, share with every member of the organisation once enrichment completes. Default `false` (private).    |
| `published_ts`   | string    | No       | Optional ISO-8601 publication timestamp. Must not be in the future.                                             |

### Important Notes

* Upload is **asynchronous**. A successful response always returns `status="processing"`; the document becomes searchable only after enrichment and indexing complete
* A newly uploaded document may take a short time to appear in [`bigdata_list_documents`](/mcp-reference/tools/bigdata-list-documents) while it is registered server-side
* Binary files (PDF, DOCX, PPTX, and similar) cannot be uploaded through this tool

## Data Returned

| Field            | Description                                          |
| ---------------- | ---------------------------------------------------- |
| `document_id`    | Content ID assigned at upload                        |
| `status`         | Always `processing` on success                       |
| `tags_applied`   | Final tag list applied to the document               |
| `share_with_org` | Whether the document is shared with the organisation |

## Practical Tips

### Tagging for later retrieval

Use consistent `prefix:value` tags so content is easy to find with [`bigdata_list_tags`](/mcp-reference/tools/bigdata-list-tags) and tag-scoped [`bigdata_search`](/mcp-reference/tools/bigdata-search) queries — for example `topic:semiconductors` or `project:earnings-q2-2025`.

### After upload

Once enrichment finishes, use [`bigdata_fetch_document`](/mcp-reference/tools/bigdata-fetch-document) to read the full document, or [`bigdata_search`](/mcp-reference/tools/bigdata-search) to retrieve it alongside other private and public content.