ClawHub

README Generator

Security checks across malware telemetry and agentic risk

Overview

This README generator appears to do the expected project inspection and README writing, with no evidence of hidden access, persistence, or data exfiltration.

Install this if you want an agent to inspect your project structure and draft a README. Review the generated README before committing it, and be aware that ordinary requests to create or write a README may activate the skill.

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 (3)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrases are generic and likely to match ordinary user requests about creating or writing a README, causing the skill to activate more broadly than intended. In an agent environment, overly broad activation can lead to unintended file reads, project scanning, and write actions such as generating or overwriting README content without the user explicitly selecting this skill.

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
90% confidence
Finding
The trigger 'create readme' is close to a generic built-in 'create' action and may shadow or intercept common assistant commands. This increases the chance of accidental invocation, especially because the skill performs repository analysis and may write to README.md, turning an ordinary request into an unanticipated autonomous workflow.

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
90% confidence
Finding
The trigger 'write readme' similarly overlaps with a generic built-in 'write' action and can capture routine requests that were not meant to invoke this specific skill. Because the skill is designed to inspect project files and produce output in the repository root, accidental activation could expose project context unnecessarily or modify files unexpectedly.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal