ClawHub
Back to skill

Security audit

Picture Book Video

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly does what it claims, but it can hand generated videos to a Douyin publishing workflow without a clear final consent step.

Review before installing. Use it only if you are comfortable with local video generation, story text being processed by remote TTS when fallback is used, and a workflow that can pass videos to a Douyin publisher. Require a separate manual confirmation before any upload.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The skill extends beyond local video generation by directing use of a separate Douyin publishing capability, which can cause external data transfer and public content posting. Without an explicit consent boundary, users may trigger publication workflows they did not intend, exposing generated media, metadata, or embedded personal content to a third-party platform.

Vague Triggers

Medium
Confidence
82% confidence
Finding
Overly broad trigger phrases increase the chance the skill activates in ordinary conversation, which is risky because the skill can read/write files, run shell commands, and chain into publishing behavior. Mis-triggering an automation skill with side effects can lead to unwanted project creation, resource consumption, or progression toward external publication.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill includes instructions to publish generated content to Douyin but does not provide user-facing privacy, consent, or external-sharing warnings. This is dangerous because children's content, voice output, titles, and story text may contain personal or sensitive information and could be uploaded publicly or to a linked account without adequate risk disclosure.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The script sends arbitrary input text to the Edge TTS service via network requests using edge_tts.Communicate, but the code provides no warning, consent check, or data-classification guard before transmitting that content off-host. In this skill's context, story text may still contain unpublished, copyrighted, or personal content, so silent exfiltration to a third-party service is a real privacy and compliance risk rather than just a theoretical concern.

Static analysis

No suspicious patterns detected.