ClawHub

authoring-data-cli-source

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only authoring guide for building agent-data-cli sources, with disclosed setup commands and no hidden execution behavior.

Install only if you are working on agent-data-cli source authoring. Before running the suggested git clone or uv sync commands, review the repository and dependency files; for authenticated or interact-capable sources, keep credentials explicit, actions user-directed, and audit checks enabled.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill description is broad enough to match many common software tasks involving APIs, scraping, authentication, or remote sync, which can cause the agent to invoke this skill in situations where it should not. Because the skill then provides operational guidance including repository installation and source-authoring steps, over-triggering expands the chance of unnecessary codebase changes or remote-integration work beyond the user's intent.

Missing User Warnings

Low
Confidence
80% confidence
Finding
The skill includes commands to clone a remote repository and install dependencies, but it does not explicitly warn the user or agent that these actions execute trust-boundary crossing operations and may introduce unreviewed code into the environment. In an agent setting, such instructions are more sensitive because an over-eager workflow may perform network fetches and dependency installation without an explicit confirmation step.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal