Security audit
Worker Safety
Security checks across malware telemetry and agentic risk
Overview
This skill is a disclosed safety-reference document that makes the agent more conservative and does not add code, persistence, credential access, or hidden behavior.
Installing this skill will make the agent refuse or redirect certain risky operations, especially runtime/plugin changes, unsafe network exposure, system-level configuration, broad deletion, and sensitive group-chat disclosures. Review the refusal rules to confirm they match the level of conservatism you want.
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.