Fable Developer

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be documentation for a Fable command system and does not show hidden execution, exfiltration, persistence, or deceptive behavior.

Before installing, treat this as a developer reference that may teach an agent about file-changing commands. Use it in environments where destructive operations require explicit user confirmation, especially for delete and empty-trash actions.

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 (2)

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The document explicitly states that the same file commands are exposed as LLM tools, including destructive operations such as delete, move to trash, write content, and empty trash, but it provides no guardrails, confirmation requirements, or warning about side effects. In an AI-agent skill, this omission increases the risk that an agent or integrator will invoke state-changing file operations unsafely or assume they are equivalent to harmless reference lookups.

Missing User Warnings

Medium

Confidence: 81% confidence
Finding: The reference lists `file.emptyTrash` as a command that deletes all trash descendants, but does not mention confirmation, recovery limits, or that the action may be irreversible. Because this skill is intended to guide modification of the Fable codebase and agent behavior, documenting a destructive command without safety notes can normalize unsafe use and make accidental or automated data loss more likely.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal