Security audit
Agent Anchor
Security checks across malware telemetry and agentic risk
Overview
The skill describes a useful local recovery dashboard, but users are told to run an unspecified external repo while it promises automatic agent-state persistence and one-click cron fixes without enough safety detail.
Review before installing. Do not run the dashboard until the actual repository URL, version, and code are known. Confirm what agent memory is saved, whether secrets and private workflow data are excluded, how stored state can be deleted, and whether cron-job fixes show a preview and require explicit confirmation.
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
64/64 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.