Back to skill

Security audit

varg-ad-variants

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: scanners found no reported malicious behavior, but one executable script was only partly inspected and a possible external-processing disclosure issue was noted.

The verdict is benign, but confidence is medium because analysis was partial: Skillspector reports 50% coverage, with the executable script only partly inspected. Before installing, consider whether you are comfortable using a skill whose script was not fully analyzed and whether any ad media it processes may be shared with an external service.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 58)May include surrounding context.

md
node scripts/hook-swap.mjs --body master_clean.mp4 --cut 7.34 --out variants/ \

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 71)May include surrounding context.

md
ooks on `eleven_v4` (1 credit each) over the reel's own clips, swapped in with `hook-swap.mjs` in seconds, then re-captioned:

The workflow directs analysis of the user's winning ad through varg without warning that the ad will be shared with an external service.

Medium
Category
Output Handling
Confidence
89% confidence
Finding

The first workflow step instructs the agent to run video_analyze on the winning ad using the external varg service. Although the skill names the service and credit cost, it does not warn that the user's media will be submitted for third-party processing, which matters for confidential or unreleased advertising assets.

Content

No source excerpt is available for this finding.