ClawHub
Back to skill

Security audit

Discord聊天

Security checks across malware telemetry and agentic risk

Overview

This Discord messaging skill is mostly understandable, but it asks for broad local powers and documents destructive Discord actions without enough scoping or confirmation safeguards.

Review this skill before installing. Use it only for explicit Discord tasks, avoid granting exec/write unless your agent platform requires them for this exact workflow, and require manual confirmation before edits, deletions, bulk cleanup, or any action using MANAGE_MESSAGES permissions.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The skill primarily presents itself as a Discord messaging interface, but later claims broader file handling, API integration, and command execution capabilities. This mismatch can mislead an agent or operator into granting or invoking unnecessarily powerful behaviors, increasing the chance of unintended file access, external calls, or shell execution outside the intended scope.

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The troubleshooting section references command-line execution and exec capability even though the rest of the skill describes Discord messaging workflows. This inconsistency can cause an agent to believe shell execution is expected or permitted, which broadens the attack surface and may lead to unsafe command invocation from unrelated user prompts.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation wording is overly broad, covering generic messaging, notifications, email/SMS, and communication integration use cases despite the skill being named and documented for Discord chat. Overbroad routing increases the risk that the skill is invoked in unrelated contexts, causing unintended data disclosure, misrouted communications, or misuse of message actions on the wrong platform.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill documents message deletion as a routine operation without requiring confirmation, warning about irreversibility, or distinguishing between the bot's own messages and others. In a communication-management context, destructive actions are especially risky because an agent may delete important records, moderation evidence, or user content based on ambiguous prompts.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.