saas-decomposer

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only SaaS analysis skill whose web crawling, memory references, and event output are disclosed and aligned with its purpose.

Safe to install for public SaaS research. Use it with explicit SaaS targets, avoid private or logged-in pages, review local memory files if present, and check generated event files before letting downstream planning skills rely on them.

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: 92% confidence
Finding: The trigger description uses broad phrases like 'decompose', 'build this service with AI', and an open-ended 'etc.', which makes activation scope ambiguous and easier to invoke unintentionally. In an agent skill that performs web crawling, comparative analysis, and roadmap generation, overbroad activation can cause the wrong skill to run on unrelated requests, leading to unintended external fetches, data processing, or misleading business/security decisions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal