AgentAudit: An Open, Extensible Framework for Full-Lifecycle Trust Evaluation of AI Agents

  • 类型:arxiv
  • 标识:2609.09875
  • 链接:http://arxiv.org/abs/2609.09875v1
  • 主分类:evaluation
  • 形态:benchmark
  • TLDR:Existing evaluation frameworks mostly assess only one part of AI agents, such as task completion (AgentBench) or security robustness (AgentDojo, ASB), rather than the complete pipeline of planning, tool selection, tool execution, memory and reasoning. Failures can occur at any stage, yet existing benchmarks rarely identify their precise source. AgentAudit evaluates the entire execution trace across ten capability, grounding, security and behavioural dimensions, namely instruction integrity, planner, memory, tool selection, tool invocation, tool correctness, alignment, tool faithfulness, securi
  • 副分类:agent
  • 待LLM分类:否
  • 来源文件
  • /inbox/tom/_candidates/2026-09-10-agent-rag-longcontext-candidates.json