OBLIVION: Workflow-Level Operational Skill Unlearning for Deployed Agents

arXiv:2608.08264 · cs.AI · Submitted 2026-08-08 · Read on arXiv

Zhengyang Shan, Xu Qian, Jiayun Xin, Kun Li, Yue Zhang, Minghui Xu

cs.AI

Submitted: 2026-08-08

Updated: 2026-08-11

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 23× in this paper · explained in 2 episodes
• Harness — 8× in this paper · explained in 2 episodes
• Trace — 6× in this paper · explained in 2 episodes
• Cross-Surface Coherent Erasure — 4× in this paper
• Indirect Prompt Injection — 3× in this paper · explained in 2 episodes
• Machine Unlearning — 3× in this paper · explained in 3 episodes
• Papers — 3× in this paper · explained in 3 episodes
• Anti-ReDerivation Guard — 1× in this paper
• Benchmarking — 1× in this paper · explained in 4 episodes
• Core — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Generator — 1× in this paper · explained in 2 episodes
• I MPACT- WEIGHTED EXPOSURE — 1× in this paper
• Model Context Protocol (MCP) — 1× in this paper · explained in 8 episodes
• Prompt Injection Attacks — 1× in this paper · explained in 2 episodes
• Task Alignment — 1× in this paper · explained in 2 episodes

Sources

• Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
• SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agents
• SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement
• How Your Credentials Are Leaked by LLM Agent Skills: An Empirical Study
• ClawGuard: A Runtime Security Framework for Tool-Augmented LLM Agents Against Indirect Prompt Injection

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems