Anki

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Anki study skill with no code execution, hidden data access, or persistence.

Safe to install for flashcard help. Share only the notes, PDFs, deck exports, or stats you want the agent to use, and specify your preferred language if you do not want English-based examples.

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

Findings (1)

Natural-Language Policy Violations

Low

Confidence: 76% confidence
Finding: The examples consistently frame learning as translation into or from English, such as '[ES→EN]' and 'See native language, produce target,' with sample answers written in English. This can reflect a language/locale assumption in the skill guidance without explicitly offering users a choice of base language or clarifying that English is only illustrative.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal