ClawHub
Back to skill

Security audit

ssgep-single-sample-expression

Security checks across malware telemetry and agentic risk

Overview

This is a transparent bioinformatics workflow skill whose broad triggers and command permissions deserve caution, but its behavior is disclosed and aligned with RNA-seq analysis.

Install only if you want this specific SSGEP/RNA-seq workflow. Before letting it run, confirm the dataset location, whether cloud processing is acceptable, and which commands or package installs will execute, especially because it has Read, Bash, and Write permissions.

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 (4)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The activation description contains very broad trigger phrases such as requests to 'do expression profiling analysis', 'package a project case', or produce 'paper three formats', which can match many ordinary bioinformatics or document-generation requests. This can cause the skill to auto-activate outside the user's explicit intent, unnecessarily exposing local data, invoking heavyweight workflows, or overriding a more appropriate skill for the task.

Natural-Language Policy Violations

Medium
Confidence
82% confidence
Finding
The skill text instructs users to interact '用中文说' and is described entirely as a Chinese-first workflow without offering a language choice. While not directly enabling code execution, this can create a misleading or coercive UX where users are steered into one language mode, increasing the chance of misunderstanding consent, workflow scope, or data-handling steps.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger phrase at this line is broad enough to match many ordinary RNA-seq or expression-analysis requests that may not specifically require this skill. Over-broad activation can cause the agent to invoke a high-permission skill (Read/Bash/Write) in contexts where a narrower or safer workflow would be more appropriate, increasing the chance of unintended file access or command execution.

Vague Triggers

Medium
Confidence
89% confidence
Finding
This trigger refers to a common task outcome rather than a precise invocation condition, making accidental or overly frequent matching likely. In the context of a research skill with Bash and Write permissions, ambiguous routing can expose users to unnecessary execution of analysis pipelines, unintended workspace modifications, or misuse on unrelated requests.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.