Security audit
省级教育课题申报Skill
Security checks across malware telemetry and agentic risk
Overview
The skill bundle is mostly coherent maintenance guidance, but its review helper grants a nested agent full sandbox-bypass access by default and may send diffs to fallback review tools.
Install only if you trust the publisher and intend to use these internal ClawHub maintenance workflows. Before running autoreview, prefer `--no-yolo` or `AUTOREVIEW_YOLO=0`, and review any fallback reviewer settings because generated diffs may leave the local toolchain. Use moderation, GitHub publishing, and Convex deploy commands only with explicit targets and dry runs where available.
SkillSpector
By NVIDIA
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
VirusTotal
66/66 vendors flagged this skill as clean.