Reversibility Marking
Reversibility marking is an AI interface design pattern that tags every agent-proposed action with an explicit signal of how easily it can be undone: fully reversible (an editable draft), partially reversible (a database row with a backup window), or irreversible (a sent email, a charged card, a deleted production record). This UX pattern prevents the common failure where agents present all actions with uniform visual weight, leading users to approve irreversible operations with the same click-through attention they give to draft edits. By showing reversibility as a first-class property next to each proposed step, the pattern calibrates user scrutiny to actual risk. It is the foundation of safe agent UX: design the undo story before you design the action.
Use Case
Essential for agent interfaces, workflow automation, and any system that performs actions on behalf of users where surfacing reversibility helps calibrate approval attention to real risk.
Examples in Wild
Use this pattern in your project
Copy this prompt to generate a production-ready implementation in Cursor, Claude Code, Lovable, or any AI coding agent.
Generate a production-ready implementation of the "Reversibility Marking" AI interface design pattern.
Pattern Description:Select actions to preview execution risk.
Related Patterns
Get new patterns by email
Weekly AI interface UX notes and resources on Substack, no spam, unsubscribe anytime.