ClawHub
Back to skill

Security audit

Discord Toolkit Free

Security checks for vulnerabilities and agentic risk

Overview

This Discord skill is mostly purpose-aligned, but its broad trigger wording and high-impact Discord actions need review before installation.

Install only if you intend to let an agent act through a Discord bot. Limit the bot's server permissions, keep the token in an environment variable, and require explicit user confirmation before deleting or editing messages, sending DMs, searching history, pinning content, or uploading local files.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The skill metadata and early description understate the feature set, while later sections document additional capabilities such as edit/delete, pinning, thread management, and message search. This mismatch can cause an agent or user to grant trust, permissions, or invoke the skill under false assumptions, increasing the chance of unintended destructive or privacy-impacting actions.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The skill states moderation and role management are unavailable in the free version, but later says moderation and roles action groups can be enabled via configuration. This inconsistency can mislead users and agents about the true privilege boundary, potentially resulting in over-permissioning or activation of higher-risk administrative operations.

Vague Triggers

High
Confidence
95% confidence
Finding
The trigger conditions are overly broad and unrelated to the skill's actual Discord-focused function, suggesting it should be used for generic coding, debugging, and deployment tasks. This can cause the agent to invoke a Discord-capable skill in inappropriate contexts, expanding message, deletion, search, or DM actions into workflows where the user did not intend Discord side effects.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill supports privacy- and integrity-impacting actions including deleting messages, sending DMs, searching message history, and reading local or remote media, but it does not provide a consolidated risk notice or consent model. In an agent setting, this increases the chance that sensitive data is accessed or actions are taken without the user understanding the privacy and moderation implications.

Static analysis

No suspicious patterns detected.