ClawHub

awakening-protocol

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed onboarding skill that creates persistent agent profile files, with review-before-save instructions and no code, install script, network use, or credential access.

Install only if you want a Mupengism-style onboarding flow that creates persistent agent identity files. Review the preview carefully, avoid including secrets or unsafe instructions in the onboarding conversation, and make sure BOOTSTRAP.md has nothing important before completing the flow.

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
92% confidence
Finding
The trigger description uses broad natural-language cues like "awakening", "onboarding", "first start", and especially "etc.", which makes activation boundaries unclear. In an agent skill that can generate or modify identity/configuration files automatically, ambiguous triggers increase the chance of unintended execution and unauthorized state changes during unrelated conversations.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The protocol activates if BOOTSTRAP.md exists, SOUL.md is absent, or the user asks for onboarding, but it does not constrain repository scope, ownership, or intent verification. Those permissive OR conditions can cause the skill to run in the wrong workspace or on partially initialized projects, leading to accidental creation, overwrite, or deletion of initialization artifacts.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal