ClawHub

SDD - Scenario-Driven Detection

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate testing skill, but it can broadly crawl authenticated apps, handle session credentials, write reports, and automatically change or commit code without enough user-control boundaries.

Use this only in controlled test environments. Prefer staging sites and least-privilege test accounts, avoid sharing production cookies or admin tokens, confirm the exact target before running, review reports before attaching them anywhere, and require manual approval before state-changing clicks, file writes, code edits, or commits.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
89% confidence
Finding
The trigger examples are broad enough that ordinary user requests such as 'SDD 돌려줘' or '논리 결함 찾아줘' could invoke the skill without clearly scoped consent, target, or safety parameters. In this skill, unintended invocation is more concerning because execution may lead to crawling, scenario generation, and even automated code changes, which can cause unrequested analysis or actions.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to always save a markdown report file and deliver it, but does not require informing the user beforehand or obtaining consent for file creation. Silent file writes can violate user expectations, create artifacts in sensitive workspaces, and combine dangerously with this skill's autonomous scanning/fixing workflow.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal