Security audit
Matapan
Security checks across malware telemetry and agentic risk
Overview
This skill gives agents a controlled workflow for making code changes in an isolated Matapan workspace and handing them to a human for review.
Before installing, understand that this skill is meant for use with a Matapan MCP server and will guide an agent to edit, test, commit, and seal code-change proposals inside Matapan workspaces. It appears appropriately scoped: humans remain responsible for grants, final apply, rejection, and teardown.
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.