ClawHub
Back to skill

Security audit

Use Self-Improving

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed read-only helper for consulting a local knowledge folder, though it may activate automatically for many project or history-related tasks.

Before installing, review what is stored under ~/.agent-knowledge because this skill may automatically consult those local records during relevant tasks. The inspected artifact does not add scripts, network behavior, writes, deletion, or privilege escalation.

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
92% confidence
Finding
The auto-trigger conditions are broad and overlap with many normal tasks, such as any clear task, mentions of prior work, project paths, or planned external operations. This can cause the skill to activate unexpectedly and inspect local knowledge sources without an explicit user request, increasing the risk of unnecessary data exposure and context contamination from unrelated historical material.

Static analysis

No suspicious patterns detected.