AgentTrust — Security Scanner for AI Skills
Security checks across malware telemetry and agentic risk
Overview
This is a documentation-only skill for sending user-selected skill or MCP content to a security scanning service, with clear privacy and payment warnings.
Installers should treat AgentTrust as an external scanning provider: only submit skill content, manifests, endpoint URLs, hashes, or wallet addresses they are comfortable sending to agenttrust.uk. Use the free endpoints first, and only approve paid USDC requests after checking the exact price and what content will be submitted.
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.