ClawHub
Back to skill

Security audit

fulcra-dashboard

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Fulcra dashboard builder that handles private data and optional publishing with consent checks, but users should review what data goes public and the use of third-party CDNs.

Before installing, be aware that this skill is meant to work with personal Fulcra data. Only approve the Fulcra data fetches you actually want, inspect the public/ directory before publishing, and avoid publishing if the included JSON/JSONL files contain private information. For maximum privacy, vendor the JavaScript and font assets locally or remove the external font/CDN dependencies before viewing or sharing sensitive dashboards.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs the agent to perform sensitive operations including querying private Fulcra data, reading local files, invoking networked CDNs, and publishing content, yet the metadata declares no explicit permissions. That mismatch weakens user awareness and policy enforcement, making it easier for the skill to access private data or exfiltrate selected artifacts without a clear, machine-readable permission boundary.

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
This public/shareable dashboard loads Alpine.js and D3 from third-party CDNs even though the page is intended to display private Fulcra data locally and can be exported for sharing. If the page is opened while connected to the internet, those external requests disclose usage metadata and create a supply-chain risk where a compromised CDN or dependency could execute attacker-controlled JavaScript in the context of sensitive local data.

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
The stylesheet imports Google Fonts from an external domain, causing clients who open the dashboard to make network requests to a third party. In a skill explicitly described as displaying private Fulcra data locally and producing shareable static output, this breaks locality/privacy expectations and can leak viewer IP addresses, access timing, referer/origin context, and usage metadata to Google or any intermediary observing the request.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The invocation guidance is broad enough that a generic request such as setting up a web app or dashboard could trigger a workflow that fetches private user data, runs shell commands, starts servers, and prepares public deployment. In this skill's context, over-broad triggering is more dangerous because the described workflow handles sensitive personal data and includes network/publication steps.

VirusTotal

52/52 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.