ClawHub
Back to skill

Security audit

ken

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed coding-style skill with no executable code, credentials, network behavior, or hidden data access, though its always-on wording is broad.

Install this only if you want the agent to default to a terse, rewrite-favoring coding style. Use 'stop ken' or 'normal mode' when you want ordinary behavior, and review any proposed deletions or rewrites carefully before accepting them.

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

High
Confidence
94% confidence
Finding
The skill declares itself active on every response and persistent until explicitly disabled, which makes its behavioral override apply far beyond a narrow, user-invoked scope. In an agent setting, this can interfere with higher-priority task routing, cause unintended behavior on unrelated prompts, and make prompt-injection or policy-conflict handling harder because the skill keeps asserting control by default.

Static analysis

No suspicious patterns detected.