Backup Before You Regret: Measuring Backup Awareness and Automating Backup in Coding Agents

Coding agents now complete tasks autonomously, modifying real-world state such as artifacts, software environments, and configurations along the way. Yet they show limited backup awareness: recognizing that an action will modify state that may later be needed, and preserving it beforehand. To evaluate backup awareness, we introduce RegretBench: 205 executable tasks grounded in regret moments reported on platforms such as X.com, GitHub, and Stack Overflow, where users regretted not preserving state they later needed. Across six frontier models, agents complete 86.1% of tasks on average yet back up before these moments in only 15.2%, showing that task completion does not imply backup awareness. Because it is hard to predict which changes will be regretted, we propose BackUpper, a harness plug-in backup layer that checks every tool call and preserves affected state before modification. It combines rule-based strategies for common destructive operations with a lightweight subagent for complex cases. BackUpper's restored state reaches 92.2% average similarity to the original, and every regret-moment action is covered. Achieving this costs only $0.014 per task, with 7× fewer tokens than main-agent self-backup. When users state backup preferences, BackUpper narrows the backup scope and achieves 84.5% F1. It can also add new rule-based strategies through self-improvement to reduce subagent calls over time. These results show that agents can keep actions reversible at low cost while respecting users' preferences.