Back to skill

Security audit

varg-character-cringe

Security checks for vulnerabilities and agentic risk

Overview

The scanners returned a benign verdict with medium confidence: they found an unpinned `npx skills` installer and a possible disclosure gap around sending media to external services, while reporting no AIG findings; coverage was partial because a local reference could not be resolved.

The verdict is benign, but benign does not mean risk-free. Before installing, consider that the documented installer is unpinned and that media used in the video workflow may go to external services. Avoid submitting sensitive or identifiable media unless you are comfortable with that processing. The scan was partial because a local reference could not be resolved.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The installation example invokes an unpinned npm package through npx, allowing newly fetched package code to execute with the user's privileges. A compromised future release could therefore affect users following these instructions; this is an installation supply-chain risk, not an MCP-server reference as the finding claims.

Content

No source excerpt is available for this finding.

The video workflow uses user footage with external generation services without warning about third-party media processing.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

This workflow step directs the use of real footage and character images with external motion-control or recasting services, rather than merely listing available tools. The warnings about owning or licensing footage address usage rights but do not disclose that the media will be sent to third-party services for processing.

Content

No source excerpt is available for this finding.