Browse Ai
Security checks across malware telemetry and agentic risk
Overview
This Browse AI skill is a disclosed Membrane-based integration, but installing it lets an agent use authenticated Browse AI account actions.
Install only if you are comfortable connecting Browse AI through Membrane and letting the agent act within that account. Prefer named Membrane actions, and require the agent to show the action or endpoint, method, and payload before bulk tasks, cookie updates, account/admin changes, invoice or API-key access, or any POST, PUT, PATCH, or DELETE proxy request.
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.