WeClawArena: An Auditable Sandbox and Benchmark for Cross-User Agents Collaboration and Security in Human-Centered Agent Networks

• 类型：arxiv
• 标识：2608.03499
• 链接：https://arxiv.org/abs/2608.03499
• 主分类：agent
• 形态：benchmark
• 被引：0
• 被引来源：Semantic Scholar
• S2被引：0
• 影响力被引：0
• TLDR：WeClawArena is introduced, an auditable benchmark and runtime sandbox for multi-party owned-agent collaboration over personal workspaces and audits attack success from bounded runtime evidence, supporting diagnosis of task breakdown, privacy leakage, poisoned evidence, and invalid authority paths.
• 副分类：evaluation
• 待LLM分类：否
• 标题中文：WeClawArena：人本 Agent 网络中跨用户 Agent 协作与安全的可审计沙箱与基准
• TLDR中文：本文提出 WeClawArena，一个面向个人工作空间多参与方 owned-agent 协作的可审计基准与运行时沙盒，基于有界运行时证据审计攻击成功情况，支持任务分解失败、隐私泄露、证据投毒以及权限路径失效等问题的诊断。
• 来源文件：
• /inbox/tom/_candidates/2026-08-11-agent-rag-longcontext-candidates.json
• [S2 enrich]