SWARM Safety
Security checks across malware telemetry and agentic risk
Overview
This is a coherent local research skill for multi-agent safety simulations, with normal caution needed for its external Python package and optional localhost API.
Install only from the expected PyPI package or GitHub repository, consider pinning the version, and enable extras only when needed. If running the API, keep it bound to 127.0.0.1, do not expose it to public networks without authentication and firewall controls, and avoid placing real secrets, credentials, or personal data in scenarios.
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.