Skip to content

Files

File Management

Hypersigil includes a comprehensive file management system that enables multimodal AI interactions by supporting file uploads, organization, and integration with prompt executions. This system allows you to work with images, documents, and other file types as part of your prompt testing workflow.

File Upload and Storage

The file management system provides a centralized location for uploading and organizing files that can be used across your prompt testing activities. Files are securely stored using base64 encoding and can be easily accessed through the intuitive file management interface.

Supported File Types

  • Image Files: PNG, JPEG, GIF, and other common image formats for visual analysis
  • Document Files: PDF files for document analysis and content extraction
  • Text Files: Various text formats for content analysis and processing
  • Other Formats: Extensible support for additional file types as needed

Upload Process

The file upload system supports both single and multiple file uploads with real-time progress tracking. When uploading files, you can:

  • Add Display Names: Provide custom names for better organization
  • Add Descriptions: Include detailed descriptions for context
  • Apply Tags: Organize files with custom tags for easy categorization
  • Monitor Progress: Track upload status with visual progress indicators

Tagging System

Files can be organized using a flexible tagging system that allows you to:

  • Create Custom Tags: Define tags that match your organizational needs
  • Batch Tagging: Apply tags to multiple files simultaneously
  • Visual Tag Display: Tags are displayed as intuitive badges in the file list
  • Tag-Based Filtering: Filter files by specific tags for quick access

Search and Filtering

The file management interface includes powerful search and filtering capabilities:

  • Debounced Search: Real-time search across file names and metadata
  • Sort Options: Sort by creation date, name, size, or file type
  • Pagination: Efficient handling of large file collections
  • Advanced Filtering: Filter by file type, size, or other attributes

File Integration with Prompts

Multimodal Prompt Support

Files can be attached to prompt executions when prompts are configured to accept file uploads. This enables sophisticated testing scenarios such as:

  • Image Analysis: Test how your prompts analyze and describe visual content
  • Document Processing: Evaluate prompt performance with PDF documents
  • Mixed Media Testing: Combine text instructions with visual or document context

File Selection in Executions

When scheduling executions for prompts that support file uploads:

  • Optional Attachment: File selection is optional and doesn't complicate the workflow
  • File Selector Component: Intuitive autocomplete interface for file selection
  • Provider Compatibility: Automatic handling of provider-specific file processing
  • Type Safety: Full validation and error handling for file attachments

File Viewing and Management

Built-in File Viewer

The system includes a comprehensive file viewer that supports:

  • PDF Viewing: Full-screen PDF viewing with navigation controls
  • Image Display: High-quality image viewing with zoom capabilities
  • Download Support: Direct file download with proper headers and streaming

File Management Operations

  • File Details: View comprehensive metadata including size, type, and creation date
  • File Deletion: Remove files with confirmation dialogs to prevent accidental deletion
  • File Download: Download files with proper binary streaming and headers
  • Bulk Operations: Perform operations on multiple files simultaneously

Use Cases and Examples

Image Analysis Testing

Upload a collection of product images and test how your prompts analyze visual content, identify features, or generate descriptions. This is particularly valuable for e-commerce, content moderation, or visual search applications.

Document Processing Workflows

Upload PDF documents such as contracts, reports, or manuals to test how your prompts extract information, summarize content, or answer questions about document contents.

Mixed Media Scenarios

Combine text-based test data with relevant images or documents to create comprehensive testing scenarios that mirror real-world usage patterns.