Questions
- Can I use sensitive data with LibreChat?
- How do I upload a file for the AI to reference in LibreChat?
- What file types can I upload to LibreChat?
- How do I manage or delete uploaded files?
- What is the Artifacts feature and how do I use it?
- How do I get LibreChat to generate a diagram, code block, or HTML preview?
Environment
This article applies to Bowdoin College faculty, staff, and students using LibreChat in a web browser. LibreChat is approved for use with sensitive data under Bowdoin's data classification policy. Data entered into LibreChat is not used to train large language models.
- Who: All current Bowdoin LibreChat users
- What: File upload (RAG), File Search, and Artifacts features in LibreChat
Data safety: LibreChat is approved for use with sensitive data under Bowdoin's data classification policy. Always follow any additional data handling requirements specific to your research, department, or role.
Resolution
Upload a file for the AI to reference
LibreChat's file upload feature lets you share documents with the AI so it can reference their contents when generating responses. This is useful for summarising research papers, extracting key information from reports, reviewing code, or analysing data tables.
- In the message field at the bottom of the LibreChat window, click the Attach Files icon (paperclip) on the left side.

- Select the file source from the upload options shown.

- Once uploaded, your file appears just above the message field.

- Type your prompt and send — the AI will draw on the file contents when responding. This uses Retrieval-Augmented Generation (RAG), where the file is queried alongside the language model to produce a grounded response.
Manage uploaded files
- Open the right-hand panel and select Attach Files from the panel drop-down.

- Click Manage Files to open the file management window, where you can view and delete previously uploaded files.

Enable File Search for an agent
When creating or editing an agent, you can enable File Search to allow the agent to search across a set of uploaded documents. This is useful for agents that serve as a knowledge base or policy reference. To enable it, open the Agent Builder, check Enable File Search in the Capabilities section, then upload your documents after the agent is created.

Use the Artifacts feature
The Artifacts feature presents AI-generated outputs — such as code, HTML, diagrams, and formatted documents — in a separate interactive panel alongside your chat. This lets you preview and interact with structured content without it cluttering the conversation.
- In the message field, click the Tools button (slider icon) on the left side.

- Select Artifacts from the options shown.

- With Artifacts enabled, send a prompt that produces structured output — for example, "Write a Python function that sorts a list" or "Create an HTML page with a two-column layout."
- The Artifacts panel opens automatically alongside the chat, showing the generated output. You can preview, copy, or further edit the content directly in the panel.
Tip: Some models work better with Artifacts than others. If the Artifacts panel does not appear or the output looks incomplete, try a different model. GPT-5 and Claude Sonnet 4 tend to produce reliable Artifact outputs.
Video walkthrough
Additional Help
If you need further assistance, you have several options:
- Bowdoin Bot: Chat with Bowdoin Bot directly from any KB page for instant answers.
- Phone: Call the Bowdoin College Service Desk at (207) 725-3030.
- In person: Visit the Tech Hub in Smith Union during business hours.
- Submit a ticket: Request assistance through the Service Catalog.