Skip to content

Conversation

@philipph-askui
Copy link
Contributor

Summary:
introduces a comprehensive system prompting framework for the Vision Agent's act() command, nudging users to write better system prompts and hence to customize agent behavior for their specific applications and UI contexts.

Key Changes:

  • feat: introduces a Five-part prompt structure: System Capabilities, Device Information, UI Information, Report Format, Additional Rules
  • feat; adds ActSystemPrompt Structure to reflect new prompt structure
  • feat: adds prompt factory functions for faster prompt creation
  • feat: adds default prompt parts for common use cases (Desktop Device, Android Device, Md Report)
  • chore!: moves existing system prompts from askui.prompts.system to askui.prompts.act_prompts as they are only meant for use with the act command
  • chore!: moves existing get prompts from askui.models.shared.prompts to askui.prompts.get_prompts

@philipph-askui philipph-askui changed the title Introduces novel prompting paradigm with 5-fold system prompt feat: novel prompting paradigm with 5-fold system prompt Dec 10, 2025
@philipph-askui philipph-askui self-assigned this Dec 10, 2025
@philipph-askui philipph-askui changed the title feat: novel prompting paradigm with 5-fold system prompt feat: novel prompting paradigm with 6-fold system prompt Jan 15, 2026
@philipph-askui philipph-askui marked this pull request as ready for review January 15, 2026 06:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants