ClawHub
Back to skill

Security audit

aida

Security checks across malware telemetry and agentic risk

Overview

This skill is not malicious, but it can invoke authenticated smart-building control and optimization actions without clear safety boundaries or confirmation.

Install only if you trust the configured AIDA endpoint and can provide a least-privilege API token. Treat control and optimization intents as live building operations, require human approval before running them, and verify results because the code can return success-like fallback messages even when an API call fails.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (3)

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The skill exposes remote-control, optimization, and diagnostics actions that directly trigger outbound requests to an external AIDA service, but there is no authentication, authorization, user confirmation, or scope restriction in the skill itself. In an agent environment, any actor able to invoke these intents could cause real-world or operational actions through the external API, making the lack of guardrails dangerous even though the code does not itself reveal the downstream effects.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill enables authenticated control and optimization of building systems such as lights, shades, and HVAC, but the description and examples do not warn users that these actions can cause real-world operational changes. This omission increases the risk of unintended or socially engineered commands being issued without adequate user awareness, especially in safety- or business-critical building environments.

Unpinned Dependencies

Low
Category
Supply Chain
Content
"version": "1.1.0",
  "type": "module",
  "dependencies": {
    "node-fetch": "^3.3.2"
  }
}
Confidence
92% confidence
Finding
"node-fetch": "^3.3.2"

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.env_credential_access

Environment variable access combined with network send.

Critical
Code
suspicious.env_credential_access
Location
skills/aida/index.js:3