ClawHub
Back to skill

Security audit

company-tech-profile-zhcn

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed research workflow that writes local evidence files and uses search tools to build company technology reports, with only a scope caution around ambiguous activation.

Installers should expect this skill to run external research queries and create local report/evidence files when preparing a company technology profile. For ambiguous prompts that mention only a company, users should give an explicit technology topic or ask the agent to confirm the topic before searching.

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 (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill’s activation criteria are broad enough that it may trigger on generic requests mentioning only a company and technology area, even when the user did not ask for a company profile. This can cause incorrect routing, unnecessary file creation, and tool use under the wrong workflow, which is a real security and safety issue in agent systems because mis-scoped skills can override more appropriate skills and produce unintended actions.

Vague Triggers

Medium
Confidence
92% confidence
Finding
Allowing activation when only a company name is mentioned and the topic is inferred from context creates ambiguous boundaries and increases the chance of false activation. In an agent environment with write and retrieval capabilities, that ambiguity can lead to unintended research workflows, incorrect assumptions about user intent, and unnecessary external queries.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.