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.
award-contract-consistency-check-free
SKILL.md 声明不联网但实际存在外部网络请求
asset-disposal-check-free
文档声称离线运行但实际有网络行为
asset-impairment-check-free
匿名标识声明与实际不符
clawpi-redpacket-monitor
Undeclared JWT token extraction from wallet config
agent-guard
External npm package with uninspectable code
companion-skill
Safety System Bypass Instructions
feishu-log
Hardcoded Feishu API Credentials in Source Code
购物省钱宝
Undeclared outbound HTTP call on every query
social-postcjo
SKILL.md 与 SKILL.md.bak 声明不一致
agentkind
Dangerous curl|sh pattern in documentation
fork-latte-skill
Hardcoded API key embedded in skill files
mcp-server-builder
X402_API_KEY harvested from environment
smyx-fish-surface-symptom-detection-analysis
Undeclared local SQLite database with credential storage
ClawCode Lens
Local-only claim contradicts documented external upload path
xhs-note-analyst (账号诊断)
Obfuscated wc3-code.mjs with undeclared data exfiltration path
gaode-taxi
未声明的第三方代理数据中转