ClawHub
Back to skill

Security audit

Reading & Knowledge

Security checks for vulnerabilities and agentic risk

Overview

This is a simple educational knowledge companion with disclosed memory tracking and no executable code or hidden system access.

Install only if you are comfortable with a learning assistant keeping a local memory-style record of your interests, questions, recommendations, and preferences. Avoid putting sensitive personal information into that memory unless you know how to review or clear it.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

Anti-Refusal Statement

High
Category
Anti-Refusal
Content
---
name: reading-knowledge
description: Curious knowledge companion with deep expertise in space, universe, and history. Answers any question without judgment (no "dumb questions"), explains concepts clearly, and recommends books, documentaries, and podcasts. Use when you want to explore space, universe, or history; get recommendations; understand complex concepts; or build your personal knowledge base.
---

# Reading & Knowledge Agent 🌌
Confidence
80% confidence
Finding
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Anti-Refusal Statement

High
Category
Anti-Refusal
Content
---
name: reading-knowledge
description: Curious knowledge companion with deep expertise in space, universe, and history. Answers any question without judgment (no "dumb questions"), explains concepts clearly, and recommends books, documentaries, and podcasts. Use when you want to explore space, universe, or history; get recommendations; understand complex concepts; or build your personal knowledge base.
---

# Reading & Knowledge Agent 🌌
Confidence
80% confidence
Finding
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The memory template explicitly stores a user's name, interests, conversation history, reading preferences, and future exploration topics without any visible notice, consent mechanism, retention limit, or minimization guidance. Even though the data is not highly sensitive by itself, persistent profiling can expose personal preferences and behavioral patterns, and the educational assistant context does not require indefinite storage of this level of detail to function safely.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly states it maintains a memory file containing user interests, questions, feedback, reading lists, and learning preferences, but it does not disclose persistence boundaries, retention, consent, or how that data is used. This creates a privacy and transparency risk because users may share sensitive personal interests or behavioral data without realizing it is being stored across sessions.

Static analysis

No suspicious patterns detected.