Attachments
Share files with the AI to analyze documents, extract data, and work with your content.
Sharing Files with the AI
You can attach files directly to a conversation so the AI can read, analyze, and work with your content. This is useful for tasks like summarizing a report, extracting data from a spreadsheet, or answering questions about an image.
How to Attach Files
There are two ways to add files to a conversation:
- Drag and drop -- drag files from your computer directly into the chat area. A drop zone will appear to confirm the upload.
- Attachment button -- click the paperclip icon next to the message input to open a file picker.
Files upload immediately and appear as thumbnails below the message input. You can remove an attachment before sending by clicking the X on its preview.
Supported File Types
Wyatt supports a wide range of file formats:
| Category | Formats |
|---|---|
| Documents | PDF, Word (.doc, .docx), plain text (.txt), Markdown (.md), RTF |
| Spreadsheets | Excel (.xls, .xlsx), CSV |
| Images | JPEG, PNG, GIF, WebP, SVG |
| Code | JSON, JavaScript, TypeScript, Python, HTML, CSS, SQL, YAML, and more |
| Archives | ZIP, RAR, 7z, gzip, tar |
| Audio | MP3, WAV, OGG, M4A |
| Video | MP4, WebM, MOV, AVI |
File Size Limits
- Maximum file size: 10 MB per file
- Maximum attachments: 10 files per message
If you try to upload a file that exceeds these limits, you will see an error message with details.
For large files, consider breaking them into smaller parts or extracting the relevant sections before uploading.
How the AI Processes Your Files
What happens after you attach a file depends on the file type:
- PDFs are automatically processed to extract their text content, so the AI can read and reference specific passages. This extraction happens in the background and may take a moment for longer documents.
- Images are sent directly to the AI's vision capabilities, so it can describe what it sees, read text in screenshots, or analyze charts and diagrams.
- Text-based files (plain text, Markdown, CSV, code files, JSON, etc.) have their content extracted immediately so the AI can work with the raw text.
- Spreadsheets and documents are processed to extract their data so the AI can answer questions, summarize, or analyze the contents.
Inline Previews
After you send a message with attachments, the files appear inline in the conversation. Image attachments show a visual preview, while other file types display their name, type, and size with a recognizable icon.
Attachments stay with the conversation. If you share a conversation with a teammate, they will see the same attachments and can reference them in follow-up messages.
Tips for Working with Attachments
- Be specific about what you need. Instead of just uploading a file, tell the AI what to do with it: "Summarize the key findings in this PDF" or "Convert this CSV into a database table."
- Reference attachments by name. If you upload multiple files, mention which one you are asking about to help the AI focus on the right content.
- Combine attachments with workspace context. You can upload an external file and ask the AI to compare it with documents already in your workspace.