ClawHub
Back to skill

Security audit

skill-consolidator

Security checks for vulnerabilities and agentic risk

Overview

This skill locally scans installed agent skills and rules to report conflicts, with its broad scan behavior disclosed and cleanup changes gated on user confirmation.

Install this only if you are comfortable with it reading your local AI-agent skill, rule, and instruction files to create conflict reports. Use `--no-write` if you want stdout-only reporting, and review affected paths carefully before approving any cleanup or merge action.

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
90% confidence
Finding
The trigger conditions are broad enough to auto-invoke during ordinary discussion about skills, conflicts, or installation planning, which can cause the agent to begin filesystem scanning without a clearly scoped user request. In this skill, invocation leads directly to running a local Node-based scanner over many user and project directories, so accidental activation increases privacy exposure and unnecessary tool execution risk.

Static analysis

No suspicious patterns detected.