jhammant/agent-drift

Stress-test AI agents for goal drift and system prompt violations. Inspired by 'Asymmetric Goal Drift in Coding Agents' (ICLR 2026).

语言：HTML · GitHub 星标：9 · Fork：0

话题：agents · ai-safety · alignment · goal-drift · llm · red-teaming · security · testing

在 GitHub 上查看原仓库