ClawHub
Back to skill

Security audit

Aloudata CAN SKILLS - analysis-report

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only report-generation orchestration skill with no hidden installer, persistence, or direct data-access code.

Install only if you trust the downstream metric-query, anomaly-detection, metric-attribution, forecast-simulation, and document-conversion skills it may call. Treat generated reports and intermediate results as potentially sensitive business data, especially when asking for broad business summaries.

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 (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger specification contains many broad, everyday phrases such as '出一份报告', '汇报', '工作总结', and '帮我做个报告', which can cause the skill to activate for vague or adjacent requests that were not intended for this orchestration flow. In an agent system, over-broad invocation can route users into unnecessary multi-step analysis, increase data access surface area through delegated skills, and create confusing or privacy-impacting behavior if sensitive business data is queried without sufficiently specific user intent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.