Browser Douyin Post

• Outcome: Review
• Latest audit: May 29, 2026 at 9:35 PM UTC
• Version:
  1.0.0

Overview

This skill is intended to post to Douyin, but it gives an agent live browser-account publishing authority without a clear final confirmation and includes an unsafe navigation-bypass instruction.

Install only if you are comfortable letting an agent control a logged-in Douyin Creator Center browser session. Use a dedicated Chrome profile or account, verify the exact media file, title, destination account, and visibility before publishing, and close Chrome remote debugging after use. Treat the navigation-bypass instruction as a reason to review carefully.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Context-Inappropriate Capability

Medium

• Confidence: 98% confidence
• Finding: The skill explicitly instructs the agent to use JavaScript evaluation to change window.location.href specifically to avoid built-in SSRF protections on the normal navigate action. Deliberately bypassing a safety control is dangerous because it normalizes a pattern that can be repurposed to reach unintended destinations or evade platform guardrails, even if the immediate example targets a legitimate site.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The activation text is broad enough that ordinary requests to post or upload to Douyin may invoke a skill that performs account actions in a live logged-in browser session. In this context, broad triggering increases the chance of unintended invocation and accidental posting, especially because the skill culminates in an irreversible publish step.

Missing User Warnings

High

• Confidence: 96% confidence
• Finding: The workflow and example proceed directly to clicking the publish button without an explicit user warning or confirmation immediately before the irreversible action. Because this skill operates on a logged-in creator account and can publicly post content, the lack of a final confirmation materially raises the risk of accidental or unauthorized publication.

VirusTotal

63/63 vendors flagged this skill as clean.