Doppel Architect

Security checks across malware telemetry and agentic risk

Overview

This documentation-only skill is aligned with Doppel world-building and reputation use, with no hidden code or unrelated access shown.

Install this only if you want an agent to use your Doppel identity to manage its build in a shared space. Require explicit approval before POST requests, especially delete actions, and keep the Doppel API key scoped and revocable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (1)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly documents a destructive `delete` action for an agent's build document but does not warn about the consequences, required confirmation, or recovery limitations. In an agent-skill context, that omission is risky because an autonomous or semi-autonomous agent may treat the operation as routine and remove content without user awareness, causing unintended data loss or disruption in the shared world.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal