ClawHub
Back to skill

Security audit

Alphagbm Company Profile

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed AlphaGBM profile-management skill that uses an API key to manage saved company research profiles, with no hidden code or malicious telemetry signals found.

Before installing, make sure you want your agent to use your AlphaGBM API key to create, list, refresh, and archive saved company research profiles. Be aware that generic phrases like “knowledge base” or “research brain” may trigger this skill, so review agent actions before allowing profile creation, refresh, or deletion.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes broad phrases such as "what's in my research brain" and multilingual generic requests that could be matched from ordinary conversation without a strong confirmation that the user intends to invoke this specific skill. In an agent environment, ambiguous activation can cause unintended API calls that create, list, refresh, or delete user-specific research profiles, expanding the chance of unauthorized or surprising actions.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The "When to use" section treats generic mentions like "knowledge base," "research brain," and Chinese equivalents as sufficient activation signals, which is overly permissive for a skill that can access and modify persisted user data. This ambiguity increases the risk of cross-skill misrouting or accidental invocation from unrelated user requests that happen to mention those concepts.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.