ClawHub
Back to skill

Security audit

Morning Report

Security checks across malware telemetry and agentic risk

Overview

This skill generates a public web-sourced game-industry morning report and sends it to a specified DingTalk channel, which is disclosed and aligned with its purpose.

Before installing or scheduling it, confirm that DingTalk target 2735046220840628 is the intended channel and that automatic daily posting is desired. Use a narrowly permissioned bot/account, and avoid enabling broad triggers if users may ask for generic daily reports that should remain private.

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
92% confidence
Finding
The trigger description includes broad phrases like “晨报”, “日报”, and “行业简报”, which can cause the skill to activate on ambiguous user requests that were not meant to invoke this workflow. Because the skill performs multi-step web collection and is later configured to send results externally, accidental invocation increases the chance of unintended data gathering and downstream actions.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill explicitly instructs automatic outbound delivery of the generated report to a fixed DingTalk target without a user-facing warning or confirmation step. This is dangerous because it can exfiltrate collected or synthesized content to an external destination automatically, especially when combined with broad invocation language and scheduled execution.

Ssd 3

Medium
Confidence
98% confidence
Finding
The workflow contains natural-language instructions to transmit the completed report to a fixed external DingTalk channel, which is a direct outbound data transfer behavior. This is risky because any collected information, generated content, or accidentally included sensitive context could be sent off-platform without runtime validation, recipient review, or least-privilege controls.

VirusTotal

50/50 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.