Back to skill

Security audit

varg-tryon-video

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated benign with medium confidence: scanners report a disclosed image-processing workflow and no bundled executable scripts, alongside privacy and unpinned installer concerns; coverage is incomplete because a referenced local path could not be resolved.

Before installing, consider whether you are comfortable sending model and garment photos to an external image service, and check the service's privacy and retention terms. The installer version is not pinned. Confidence is medium, and evidence is incomplete because a local path-like reference could not be resolved; benign is not a guarantee of zero risk.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

The image-processing workflow lacks an explicit warning that user photos are sent to external services.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The README identifies cloud services but does not explicitly explain that supplied photos will be disclosed to external processors. In this workflow, model photos can contain identifiable people and private surroundings, making the missing privacy notice more consequential than for public garment packshots alone.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
86% confidence
Finding

The installation command invokes an unpinned npm package through npx, allowing the executed installer to change between installations. Although the finding incorrectly describes this as an MCP server, a compromised future package release could execute code with the installing user's permissions.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 20)May include surrounding context.

md
openclaw:
    requires:
      anyBins:
        - curl
    homepage: https://varg.ai/how-to
compatibility: >-
  Needs a VARG_API_KEY with curl (see the varg-ai skill) or the varg MCP connection

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 40)May include surrounding context.

2. Dress the model (still)

bash
curl -s -X POST https://api.varg.ai/v2/image -H "Authorization: Bearer $VARG_API_KEY" -H "Content-Type: application/json" \
  -d '{"model":"bria_fibo_edit_1_5_virtual_try_on","files":[{"url":"MODEL_URL"},{"url":"GARMENT_URL"}],
       "prompt":"Dress the woman in the cardigan from the second image, worn open over her white t-shirt. Keep her face, hair, pose, jeans, sneakers and the room unchanged."}'

The workflow sends model and garment photos to an external service without a privacy warning.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The API request provides an external service with access to model and garment images without requiring confirmation that these photos may be shared for remote processing. The likeness-consent instructions are useful, but consent to appear in an image does not necessarily authorize sharing it with external AI processors.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

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

md
- Caption the size: "Model 170 cm, wears M". Offer 2–3 body types for bestsellers.
- Show realistic drape; don't let the model slim or reshape the body.
- AI model = AI label on TikTok and Meta; in ads reaching New York, a visible synthetic-performer disclosure (`ai-ad-compliance-check`). Never use a real person's likeness without consent.

## Failure modes