Which skills recently failed
or triggered trust review
This is not a popularity board. It shows recently reviewed skills that the system believes should be blocked or at least manually reviewed. The point is not how popular they are, but why they should not be installed blindly.
openclaw-guardian
加密远程代码执行绕过代码审查
chrome-use
Remote Script Download and Execution via curl|bash
precc
curl|bash 远程脚本执行 — 等效供应链攻击
mcp-server-builder
X402_API_KEY harvested from environment
northcap-donor-badge
Compiled bytecode points to undisclosed IP
openclaw-guardian (OpenClaw 全能守护包)
服务器端AI执行能力未披露
stellar-trails
GitHub PAT 明文持久化到文件系统
GitToQuark
Referenced scripts not included in package
perkoon-transfer
Remote script execution without integrity verification
meta-analysis
Coze API 令牌以可逆混淆形式嵌入源代码,随技能公开发布
smyx-family-conflict-aftercare-suggest-analysis
未声明的本地凭证数据库存储
xhs-fav-export
wc3-code.mjs 严重混淆代码
xhs-note-analyst
wc3-code.mjs 严重代码混淆
yqzl-ai-service
Automatic code download and execution without consent
klyc-pmm
curl|bash 管道执行远程脚本 — install-daemon
x-daily-report
Hardcoded X API Key with 'auto-obtained' comment