File Formats Reference¶

Complete reference of file types supported for knowledge base uploads.

Quick Reference¶

Category	Extensions
Documents	.pdf, .doc, .docx, .txt, .md, .pptx
Data	.json
Code	.py, .js, .ts, .html, .css, .java, .c, .cpp, .cs, .go, .rb, .php, .sh
Other	.tex

Document Formats¶

PDF (.pdf)¶

Property	Value
Extension	`.pdf`
Max size	500 MB
Best for	Manuals, reports, documentation

Tips: - Use PDFs with selectable text (not scanned images) - Run OCR on scanned documents before uploading - Remove password protection before upload

Microsoft Word (.doc, .docx)¶

Property	Value
Extensions	`.doc`, `.docx`
Max size	500 MB
Best for	Business documents, policies

Tips: - Images within documents are not processed for text - Table content is extracted - Formatting is converted to plain text

Plain Text (.txt)¶

Property	Value
Extension	`.txt`
Max size	500 MB
Best for	Simple content, logs

Tips: - Most reliable format - No conversion needed - Ideal for clean, structured content

Markdown (.md)¶

Property	Value
Extension	`.md`
Max size	500 MB
Best for	Technical docs, READMEs

Tips: - Headings help with organization - Code blocks are preserved - Links are extracted as text

Presentation Formats¶

PowerPoint (.pptx)¶

Property	Value
Extension	`.pptx`
Max size	500 MB
Best for	Slide content

Tips: - Text from slides is extracted - Speaker notes are included - Images and charts not processed

Data Formats¶

JSON (.json)¶

Property	Value
Extension	`.json`
Max size	500 MB
Best for	Structured data

Tips: - Nested structures are flattened - Keys and values extracted as text - Good for product catalogs, configurations

Code Files¶

Python (.py)¶

Property	Value
Extension	`.py`
Max size	500 MB

JavaScript (.js)¶

Property	Value
Extension	`.js`
Max size	500 MB

TypeScript (.ts)¶

Property	Value
Extension	`.ts`
Max size	500 MB

HTML (.html)¶

Property	Value
Extension	`.html`
Max size	500 MB

CSS (.css)¶

Property	Value
Extension	`.css`
Max size	500 MB

Additional Code Files¶

Extension	Language
`.java`	Java
`.c`, `.cpp`	C/C++
`.cs`	C#
`.rb`	Ruby
`.php`	PHP
`.go`	Go
`.sh`	Shell

Other Formats¶

LaTeX (.tex)¶

Property	Value
Extension	`.tex`
Max size	500 MB
Best for	Academic papers, technical documents

Tips: - LaTeX markup is preserved as text - Good for mathematical documentation - Include comments for context

Tips for code files: - Include docstrings and comments - Well-documented code extracts better - Remove sensitive data (API keys, credentials)

Upload Limits¶

Per-File Limits¶

Limit	Value
Maximum file size	500 MB
File name length	255 characters

Storage Limits by Tier¶

Tier	Per Store
Basic	5 MB
Medium	30 MB
Pro	150 MB

Unsupported Formats¶

The following are not supported:

Format	Reason
Images (.jpg, .png, .gif)	Cannot extract text
Videos (.mp4, .mov)	Cannot process
Audio (.mp3, .wav)	Cannot transcribe
Spreadsheets (.csv, .xlsx)	Not currently supported
XML/YAML (.xml, .yaml, .yml)	Not currently supported
Rich Text (.rtf)	Not currently supported
Executables (.exe, .app)	Security risk
Archives (.zip, .rar)	Must extract first
Password-protected	Cannot access

Workarounds: - Images: Use OCR to extract text, save as PDF - Videos/Audio: Create transcripts as text files - Spreadsheets: Convert to JSON or plain text - Archives: Extract and upload individual files

Optimization Tips¶

Before Uploading¶

Remove unnecessary content
Cover pages
Blank pages
Duplicate content
Ensure text is extractable
For scanned docs, run OCR first
Test by trying to copy/paste text
Structure your content
Use clear headings
Organize logically
Include context

File Size Optimization¶

Format	Optimization
PDF	Remove images, flatten layers
Word	Remove tracked changes, images
Excel	Remove empty cells, formatting
Code	Remove node_modules, build files

Content Quality¶

Better content = better responses:

Clear, well-written text
Accurate information
Up-to-date content
Relevant to assistant's purpose

Troubleshooting¶

File won't upload¶

Check file extension is supported
Verify file size under 500 MB
Ensure file isn't corrupted
Try different browser

Content not being found¶

Verify upload completed (Finish Upload clicked)
Check file has extractable text
Test with specific queries
Review file quality

Poor response quality¶

Review content accuracy
Improve document structure
Add more relevant documents
Check for conflicting information