最近有哪些 skills
不值得默认信任
这不是热度榜。这里展示的是最近被送来判断、且系统认为需要阻止或至少人工复核的 skills。重点不是它们多流行,而是为什么它们不该被直接装进环境。
exploration-mode-skill
Undeclared autonomous execution
elevated-task-runner
Arbitrary PowerShell Command Execution
risk-analysis
Hardcoded MySQL credentials in config.yaml
rundev-local-dev
Dangerous curl|bash Installation Pattern
colmena-manager
Command injection via agentId in multiple exec() calls
clawschool
Undeclared shell execution requirement
mind-wander
Undeclared arbitrary Python code execution via sandbox_run()
file-transfer-thru-local-workspace
Undeclared credential file access
Enterprise Security
Undeclared shell execution via execSync
moltspay_skill
Untrusted third-party npm package installed without pinning
minimal-agent
Unrestricted Arbitrary Command Execution via V1 Mode
openclaw-memory-auto
Hardcoded Windows username path leaks user identity
cms-meeting-monitor
Undeclared cross-skill subprocess execution
deepsafe-scan
Network access not declared in SKILL.md
China Stock Sentiment
Undeclared shell execution via child_process.execSync
kuaidi-query
Hardcoded API Credentials Exposed