Topics Covered:

• Worker Chat: Text, Image and URL interaction capabilities
• Workers Launchpad: Central hub for users (Coming soon)
• Session Management: History, context, and persistence
• File Upload & Management: Temporary vs persistent file handling
• Search & Filtering: Tag-based and semantic search

Worker Chat

The primary interaction interface where Users communicate with Universal Workers.

• Supports text, image, and URL input
• Users can view the Worker’s skills and knowledge sources.
• Displays the Worker's thinking trace (collapsed chain-of-thought) for transparency
• Chat controls:
  • Enter to send
  • Shift + Enter for new lines
  • Paste URLs inline
  • Attach files in chat (Coming soon)

Session Management

Each new conversation with a Worker creates a dedicated session with its own memory, settings, and context.

• All sessions are automatically titled (based on summary + timestamp)
• Users can:
  • Rename, search, or delete sessions
  • Enable/disable memory or skills for a specific session
  • See a list of previous sessions in the left-hand panel
• Planned: Graph-based session navigation (jump from old message threads into new sessions)

File Upload & Management

Users can upload documents to provide additional context during conversations.

• In-chat uploads are temporary (valid only during the session) (Coming soon)
• Persistent files are uploaded via the Session Settings panel, becoming part of session memory
• Supported formats: PDF, Word, images
• Files are embedded into vector memory and used for response generation when enabled

Search & Filtering

Users can easily discover and navigate content through search and tag systems.

• Tag-based filters on the Worker list (customizable by Builder)
• Semantic search in memory items, session names, and chat logs
• Planned enhancements:
  • Search by Worker capabilities (e.g., "find Workers that connect to Salesforce")
  • Combined filters (e.g., tags + activity status)

Summary

EverWorker's user experience is built around clarity, control, and context continuity. From rich, multi-modal chat to evolving session intelligence and personalized launchpads, the interface aims to make advanced AI interactions feel familiar, productive, and trustworthy - especially for non-technical users.