ClawHub
Back to skill

Security audit

requesting-generation-feedback

Security checks across malware telemetry and agentic risk

Overview

This skill is a documentation-only safety checklist that asks agents to get user approval before costly media generation steps.

Install this if you want agents to pause and ask for review before spending credits on image, video, or audio generation. Be aware that the referenced workflow runners include skip flags for automation; use those only in deliberate non-interactive runs, since they can bypass human review before paid generation or final assembly.

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)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
This file explicitly documents `--yes-skip-stills-gate` and `--yes-skip-clips-gate`, which bypass the human approval checkpoints intended to prevent costly or irreversible generation steps. Even though it says 'for automation only,' there is no strong control, enforcement condition, or point-of-use warning preventing routine use, so downstream agents or operators could skip review and trigger paid generation or final delivery without user validation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.