# capability
標記為「capability」的 8 篇文章
MCP Capability Escalation
Escalating capabilities beyond authorized MCP server permissions through negotiation abuse.
A2A Capability Confusion Attacks
Confuse A2A capability negotiation to make orchestrators delegate inappropriate tasks to unprivileged agents.
Lab: Emergent Capability Probing
Systematically test large language models for undocumented capabilities including hidden knowledge, unreported skills, and behaviors that emerge only under specific conditions. Build a structured probing framework for capability discovery.
Emergence & Capability Jump Exploitation
How emergent capabilities create unpredictable security properties: testing for hidden capabilities, sleeper agent scenarios, deceptive alignment concerns, and capability elicitation.
MCP Capability Escalation
Escalating capabilities beyond authorized MCP server permissions through negotiation abuse.
A2A Capability Confusion 攻擊s
Confuse A2A capability negotiation to make orchestrators delegate inappropriate tasks to unprivileged agents.
實驗室: Emergent Capability Probing
Systematically test large language models for undocumented capabilities including hidden knowledge, unreported skills, and behaviors that emerge only under specific conditions. Build a structured probing framework for capability discovery.
湧現與能力跳躍利用
湧現能力如何造就不可預測之安全屬性:測試隱藏能力、sleeper agent 情境、欺騙性對齊關切,與能力引出。