ClawHub
Back to skill

Security audit

book-breakdown

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent book-summary/report generator with expected file and web use, but users should be aware it may create notes and search online by default.

Install this if you want a book-note generator that can create workspace files and enrich reports with online metadata. For private manuscripts, confidential PDFs, or sensitive reading lists, ask for pure text or offline-only processing and decline HTML or web enrichment.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrases are very broad and map to common book-related requests, so the skill may activate when users did not explicitly intend to invoke it. That increases the chance of unexpected behavior such as unsolicited file generation, network access, or workflow takeover from more appropriate skills.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill defaults to creating Markdown and HTML artifacts in the workspace without requiring explicit user consent at the moment of write. Unexpected file creation can violate user expectations, clutter shared workspaces, and in some environments expose sensitive derived content to other tools or users with workspace access.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill directs the agent to perform web searches and fetch external pages for book metadata and reviews without a clear user-facing disclosure that network access will occur. If user-supplied titles, links, or partial documents are sent to external services implicitly, this can leak reading interests, confidential document context, or proprietary material.

Static analysis

No suspicious patterns detected.