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

# Upload documents

> Upload Markdown and text files to a Helix Foundry ontology so generations and loads read the names and facts they hold

Guide

Documents carry what your tables don't: who signed a contract, what was agreed, who
referred whom. Upload them to an ontology, and each load reads them into its database,
linking the names they mention to the records your tables hold.

## Upload

In the ontology's **Documents**, choose files or drop them on the page. Foundry checks
each file before it uploads it, then sends it straight to storage with its progress.

| Rule | Value |
| - | - |
| File types | Markdown (`.md`, `.markdown`) and text (`.txt`) |
| Size | Up to 256 KiB each |
| Count | At most 200 documents in the workspace |
| Name | At most 255 characters, with no slash, backslash or control character |

![The Documents page with contract.md uploaded](https://assets.helix-db.com/docs/foundry/ontology/documents-uploaded.png)

## How documents are used

* **Generations** give the agent the names of the ontology's 50 most recently uploaded
  documents, so a draft can add types for what they hold, such as a contract.
* **Loads** read every document: its sections, the names it mentions, and the facts it
  states, each with the quote it came from. What a load can't settle, such as a name
  the database doesn't hold, an ambiguous name, a quote it couldn't verify or a fact
  that conflicts with your tables, goes to **Needs review** instead of becoming a
  record.
* A document uploaded after the current load is read into that load as it arrives.

Each document's status says where it stands: **Waiting for a load**, **Queued for the
load**, **Reading**, **Read**, **Needs review**, **Duplicate** or **Couldn't be read**.

![The Documents list with contract.md read and one item needing review](https://assets.helix-db.com/docs/foundry/ontology/documents.png)

Reading documents uses AI models: a load sends each document's text to Anthropic and
OpenAI models to find its names and facts and to check them.

## Read a document

Open a document to see its text as the load read it: names underlined, quoted facts
tinted and anything that needs review marked. Beside it, **Facts**, **Entities** and
**Needs review** list what the load found, each linked to its record.

![contract.md with names underlined and facts tinted, and its two signed\_with facts listed with their quotes](https://assets.helix-db.com/docs/foundry/ontology/document.png)

Use the **Load** picker to read the document as another load saw it.

## Delete a document

Select **Delete** on a document's page. Owners and admins upload and delete documents;
members read them.

## Next steps

> **[Load data](/foundry/ontologies/load-data.md)**
>
> Write tables and documents into the database.

> **[Records and entities](/foundry/explore/records-and-entities.md)**
>
> See every place a record comes from.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.