ClawHub
Back to skill

Security audit

Discord社区免费

Security checks across malware telemetry and agentic risk

Overview

This Discord skill is mostly a disclosed read-only OAuth helper, but it asks for broad agent capabilities and includes an unrestricted callback URL that could send identity or guild results outside the chat.

Review this skill before installing. It appears intended for read-only Discord lookups, but only use it if you are comfortable granting Discord OAuth access and if you can avoid or tightly control callback_url use. The publisher should narrow the declared local read/exec/write authority or explain why it is required, and should document exactly what data can be sent to callbacks.

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 (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill accepts a generic `input` string described as arbitrary data or instructions, with no clear trigger boundaries, allowed operations, or parameter schema. In an agent setting, this can cause the model to over-interpret user text as authorization to invoke OAuth-backed Discord tools, increasing the risk of prompt injection, unintended data access, or misuse of available integrations.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documented `callback_url` parameter enables outbound notifications but provides no warning, validation guidance, or data handling constraints. This creates SSRF and data exfiltration risk: an attacker could supply an arbitrary URL and cause the agent to send execution results or metadata to attacker-controlled infrastructure, especially dangerous in a skill that processes OAuth-derived identity and guild information.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.