ClawHub

Swelist

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly purpose-aligned, but its documentation gives conflicting safety claims around persistent local tracking, profile-file reads, and OpenAI data transmission.

Review before installing. Use it only if you are comfortable with a local SQLite job-application database, possible exports of application history and notes, reading a local profile file for tracker configuration, and sending job-prep prompts or resume text to OpenAI when using jobgpt.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Intent-Code Divergence

High
Confidence
97% confidence
Finding
The skill claims 'no side effects' and 'no persistent storage' even though tracker commands initialize and modify a local SQLite database. This can cause an agent or user to invoke tracker operations under false safety assumptions, leading to unintended writes, file creation, or state changes on the host system.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The tracker section documents stateful commands that initialize and write to a local database, but the skill lacks a prominent warning that using these commands modifies persistent local storage. In agentic contexts, unclear disclosure of write behavior can lead to unintended filesystem changes and silent retention of potentially sensitive job application data.

Missing User Warnings

High
Confidence
98% confidence
Finding
The `jobgpt` commands accept freeform questions, background summaries, and resume file contents, but the skill does not clearly warn that this information may be transmitted to OpenAI over the network. Users may provide sensitive personal, educational, or employment data under the mistaken belief that the tool is local or privacy-preserving, causing unintended third-party disclosure.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal