ClawHub

Repo

Security checks across malware telemetry and agentic risk

Overview

This skill generates a CONTRIBUTING.md guide from ordinary project configuration files and does not show hidden, destructive, or data-exfiltrating behavior.

Install this if you want an agent helper for drafting CONTRIBUTING.md from a repository. Be aware that broad requests like "init" may invoke it, and review the generated file before accepting or overwriting anything.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
88% confidence
Finding
The manifest description embeds multiple generic trigger phrases such as "init", "project init", and "initialize project", which can cause the skill to activate in situations broader than the narrow CONTRIBUTING.md generation use case. Overbroad activation increases the chance of unintended routing, where the agent invokes project-scaffolding behavior on ambiguous user requests and exposes repository analysis or file-generation capabilities without sufficiently specific user intent.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The skill states it will "Match project language (English for open source, Korean for internal)" without clear user opt-in, which imposes output-language behavior based on inferred project context rather than explicit instruction. This can lead to unexpected or inappropriate content generation, especially if the inference is wrong, and may override user expectations or higher-level application language settings.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrases are broad enough that normal user requests about contribution guides or project setup could activate this skill unexpectedly. In an agent environment, ambiguous activation can cause unintended file inspection and generation actions, increasing the chance of surprising behavior or accidental overwrites if combined with weak confirmation handling.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal