Unverified50% confidenceCautionExact time
Before installing community plugins that involve network exposure (such as Webline Access), the code should be reviewed first, because most of these plugins have not been officially audited and AI agents have code execution and file system access privileges.
1
Sources
50%
Confidence
Long-term
Relevance
9/10/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe biggest security risk of allowing AI to directly invoke web capabilities is security itself; WebMCP-type protocols must have built-in permission authorization models, user confirmation mechanisms before critical operations, and tool source trustworthiness verification.72% similarUnverified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70% similarUnverifiedThe open-source model allows users and security researchers to audit the code for improper data collection or permission abuse, and allows the community to fix compatibility issues when system updates arise.68% similarUnverified创建仓库时务必设为私有，否则任何人都能看到代码，含有API密钥等敏感数据会造成泄露67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/891629API
curl https://kongchang.com/api/v1/knowledge/claims/891629MCP
get_claim(id=891629)