Security audit
Programmer Cognition
Security checks across malware telemetry and agentic risk
Overview
The inspected skills are coherent development and maintainer helpers with disclosed powerful actions and no evidence of hidden or malicious behavior.
Install only if you trust the maintainer workflow and expect the agent to operate in ClawHub or Convex projects. Review the moderation and autoreview instructions carefully: they can use existing authenticated CLIs, change server-side moderation state, publish PR proof, and run nested review with full local access unless you opt out.
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
65/65 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.