ClawHub
Back to skill

Security audit

AI大图生成器-by Digilifeform

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Datu poster and image-editing integration that sends selected text, files, or images to Datu using the user's Datu API key.

Install only if you are comfortable sending chosen text, documents, and images to Datu's external service. Store DATU_API_KEY in the host's secure credential vault, avoid uploading secrets or regulated data unless explicitly approved, and use confirmation for generic edit requests or deep research because they may send content off-platform and consume credits.

SkillSpector

By NVIDIA
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 (3)

Vague Triggers

Medium
Confidence
96% confidence
Finding
The trigger list includes very broad, everyday phrases such as “改图”, “修改图片”, and “换字”, which can cause the skill to activate in many unrelated contexts. In an agent setting, overbroad routing can send user content, files, or images to an external service unexpectedly, increasing the risk of privacy leakage, unintended actions, and user confusion.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger list includes broad, everyday phrases like '修改图片', '换字', and '调整颜色' that can cause the skill to activate for generic image-editing requests without strong user intent to invoke this specific external service. Because the skill uploads user text, documents, or images to a third-party API and uses a persisted API key, overbroad routing increases the risk of unintended data disclosure and accidental paid operations.

Ssd 3

Medium
Confidence
91% confidence
Finding
The changelog explicitly states that the user's original information should be passed directly to the external Datu service without summarization, which increases the chance of sending unnecessary sensitive natural-language data off-platform. In this skill context, the danger is elevated because the same changelog also describes external API usage and credentialed requests, making unintended data disclosure to a third party more plausible.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.