Lobster Agentic Engineering
Security checks across malware telemetry and agentic risk
Overview
This is a non-executable workflow guide for agentic engineering and does not request sensitive access, persistence, or hidden behavior.
This appears safe to install as a workflow skill. Its practical risk depends on the permissions granted to the agent while using it, especially repository write access, code execution, production systems, or credentials; keep those controlled by normal workspace policy.
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.