Unverified90% confidenceFactTime unknown
OpenClaw中最强大也最危险的工具是Execute，可以执行任何Shell命令，存在语言模型执行破坏性命令（如rm -rf）的风险
1
Sources
90%
Confidence
Long-term
Relevance
6/3/2026
First Seen
Sources
Related Entities
Related Claims
Unverified直接在宿主环境执行来自网络下载或智能体自主生成的Skill代码存在安全隐患，可能包含删除文件、访问敏感目录或发起网络攻击等危险操作66% similarUnverified使用他人打包的Skill文件存在安全风险，可能引入恶意指令64% similarUnverifiedExcessive blocking is itself a security risk because it prevents defenders from testing systems and releasing patches, while malicious actors can still use other models including open-source ones.62% similarUnverifiedAnthropic和OpenAI相继发布关于对抗性蒸馏的安全警告，指出攻击者可通过精心构造的提示词从闭源模型中提取核心能力62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/40381API
curl https://kongchang.com/api/v1/knowledge/claims/40381MCP
get_claim(id=40381)