ClawHub
Back to skill

Security audit

Aic Dashboard

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a local read-only dashboard skill, but its documentation and declared permissions are broader and less consistent than that purpose requires.

Review this skill before installing. Use it only for the AIC local dashboard, keep it bound to 127.0.0.1 unless LAN sharing is intentional, set a strong DASHBOARD_TOKEN, avoid sharing URLs containing the token, and do not grant write or broad command authority unless the publisher clarifies why a read-only dashboard needs it.

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 (3)

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The skill repeatedly markets itself as a read-only dashboard, but later advertises generic file processing, API integration, command execution, and the declared tools include read/exec/write. That mismatch can mislead users and agent frameworks into granting broader capabilities than expected, increasing the chance of unintended file modification or command execution under a 'safe dashboard' pretext.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation guidance is vague and overly broad, effectively saying to use the skill whenever the user needs related functionality. Ambiguous invocation criteria can cause an agent to select this skill in unintended contexts, exposing local files, starting services, or using exec-capable tooling more often than necessary.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation instructs users to access the dashboard with a token embedded in the URL. URL tokens are commonly leaked via browser history, screenshots, logs, proxy records, and referrer headers, so this pattern weakens the only stated access control protecting mailbox and session-status data.

Static analysis

No suspicious patterns detected.