ClawHub
Back to skill

Security audit

Cognitive Loop

Security checks for vulnerabilities and agentic risk

Overview

The available evidence shows a broadly triggered planning skill concern, but not hidden, destructive, or purpose-mismatched behavior.

Install only if you want this skill to help with broad planning or multi-step work. To keep control, invoke it explicitly for relevant tasks and ask the agent to confirm before writing memory, making persistent changes, or acting beyond the current request.

SkillSpector

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
89% confidence
Finding
The activation criteria are broad enough to trigger on many routine development or multi-step tasks, which can cause the skill to be invoked more often than the user expects. In this skill’s context, that increases the chance of handing substantial autonomy, planning authority, and memory-writing behavior to an external package before its safety has been fully validated.

Static analysis

No suspicious patterns detected.