ClawHub
Back to skill

Security audit

Fund Daily Report

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches a fund-reporting purpose, but its script contains an embedded JQData username and password and can silently use that account as a fallback data source.

Review before installing. The core report generation is understandable, but the embedded JQData credential should be removed and rotated, and any authenticated data provider should require explicit user configuration through secrets or environment variables.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding
The skill’s declared purpose is a daily fund report, but the documented behavior expands into sector flow analysis, stock-level inference, and—per the finding—use of a fallback service with hardcoded credentials. This mismatch is dangerous because users and orchestrators may invoke the skill under a narrower trust assumption while it accesses broader market data sources and potentially embedded secrets, increasing the chance of unauthorized data access or unintended execution scope.

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The file contains hard-coded JQData credentials, which exposes a third-party account secret directly in source code. Anyone with access to the skill can reuse the credentials, abuse the account, exfiltrate paid data, incur charges, or cause account suspension; the fallback to an authenticated service also contradicts the stated use of public data and increases suspicion.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The activation conditions are broad enough that the skill may trigger on generic investment-related requests without clear user intent for this specific data collection and reporting workflow. In an agent environment, over-broad activation increases the risk of unintended tool use, unnecessary external queries, and disclosure of derived financial analysis when a narrower or safer response would have sufficed.

Missing User Warnings

High
Confidence
99% confidence
Finding
Using hard-coded credentials without disclosure creates a clear secret-management vulnerability and hides networked access to a third-party account from users. In this skill context, that is more dangerous because the advertised purpose is a simple public-data report generator, so consumers would not expect embedded paid-service credentials or account-backed API usage.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.