ClawHub
Back to skill

Security audit

FreeAds 随手拍广告

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent paid ad-video generation skill, but users should understand that product images and prompts go to Atlas Cloud and the setup helper can store an API key locally.

Install only if you are comfortable sending product photos, brand/logo assets, prompts, and generated videos to Atlas Cloud or its downstream model providers and paying per generation. Use a dedicated low-balance API key where possible, avoid confidential or regulated images unless approved, and be deliberate before letting the helper script save the key permanently in shell configuration files.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The invocation examples include generic phrases like “生成广告视频” that could be triggered during ordinary conversation, increasing the chance of unintended activation. In a skill that uploads user-provided image URLs to third-party services and incurs per-use cost, accidental invocation can cause unexpected data disclosure and charges.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The README describes third-party processing by Atlas Cloud and external model providers, but it does not clearly warn users that image URLs, prompts, and generated outputs are transmitted to external cloud services. This creates a meaningful privacy and consent risk, especially because users may provide proprietary product images or internal assets without realizing they leave the local environment.

Vague Triggers

Medium
Confidence
77% confidence
Finding
The trigger phrases ('随手拍广告', '生成广告视频', '产品广告') are broad and generic, making accidental invocation plausible in normal conversation about ads or products. In a skill that may prompt for API-key setup or initiate paid media-generation workflows, unintended activation can lead to surprise actions, confusion, or unnecessary cost exposure.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The document instructs users to upload media and send prompts/images/API keys to a third-party cloud service without any privacy, retention, or data-handling warning. In this skill's context, users are expected to submit product photos and branding assets, which may include proprietary or unreleased materials, creating a realistic risk of unintended data disclosure to an external processor.

Static analysis

No suspicious patterns detected.