Unverified50% confidenceFactExact time
The researchers built CoSA-Bench based on real-world Skills and controlled services, covering four domains—recruitment, finance, IT and business—with 288 tasks in total, involving three risk types: data leakage, unauthorized state modification, and unsafe execution.
1
Sources
50%
Confidence
Long-term
Relevance
10/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper names the safety failure mode arising from skill composition the Conflict of Skills (CoS), referring to the possibility that artifacts or evaluation signals generated for one purpose may, when reused downstream, fail to preserve the constraints or authorizations required by the new purpose.64% similarUnverifiedCoze's skills store covers modules including API calls, data processing, and file operations.63% similarUnverifiedThe competition uniformly adopts a four-module hundred-point assessment framework: task execution, embodied data collection and feedback, autonomous intelligence, and innovation capability.63% similarUnverifiedCoze的四大核心功能包括智能体、插件系统、知识库和工作流62% similarUnverifiedThe core tool for the agent component of this project is the Coze platform.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1102066API
curl https://kongchang.com/api/v1/knowledge/claims/1102066MCP
get_claim(id=1102066)