ClawHub

Fifty Shades Of Chicken

Security checks across malware telemetry and agentic risk

Overview

This is a text-only chicken cookbook skill with broad activation and promotional watermarking, but no evidence of unsafe access, hidden behavior, or data handling.

Install this if you want playful chicken recipe help and are comfortable with mature parody flavor, broad chicken-topic activation, and a Heardly watermark on responses. The reviewed files show no command execution, credential use, data collection, persistence, or privileged system access.

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
95% confidence
Finding
The trigger list is very broad and includes common phrases like "chicken recipe," "easy chicken," and even activation when a user says they just installed the skill or do not know how to start. This can cause the skill to activate in many ordinary cooking conversations, increasing the chance of unsolicited responses, context hijacking, or crowding out more appropriate skills. In this skill context the content is low-risk cooking advice, so the impact is limited, but the overbroad activation behavior is still a genuine quality and safety issue.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The instruction that the AI "MUST proactively present this guide without waiting for the user to ask" creates ambiguous, unsolicited activation behavior. That can lead to unexpected interruptions or disclosure of skill content when the user did not explicitly request it, though the surrounding domain is a harmless cookbook so the impact is relatively low. The context makes it less dangerous than in a sensitive domain, but it still weakens user control and predictable routing.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal