ClawHub
Back to skill

Security audit

remember-books

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward reading-journal integration that sends user-provided book notes and queries to BlueColumn/Supabase, with privacy and API-key handling caveats but no hidden or destructive behavior found.

Install only if you are comfortable sending book notes, preferences, tags, and recall queries to BlueColumn's hosted service. Avoid storing sensitive personal details unless you trust the provider's retention and privacy practices, and keep the BLUECOLUMN_API_KEY in a secure environment variable or secret store.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill explicitly instructs users to send reading-history content, personal preferences, and reflective notes to a remote API, but it provides no privacy notice, data-handling warning, or guidance on minimizing sensitive content. Because reading journals can reveal personal interests, beliefs, health, politics, or work context, omission of a disclosure meaningfully increases the risk of unintended data exposure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

The example performs an outbound POST to a remote Supabase endpoint and includes detailed reading-journal content in the request body. In the context of a personal memory skill, this is expected functionality, but it still constitutes external transmission of potentially sensitive personal data and becomes dangerous without transparency, consent, and data-minimization guidance.

Content

Scanner excerpt · SKILL.md (reported line 15)May include surrounding context.

Store each finished book with the one or two ideas that changed how you think.

bash
curl -X POST https://xkjkwqbfvkswwdmbtndo.supabase.co/functions/v1/agent-remember \
  -H "Authorization: Bearer $BLUECOLUMN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "The Mom Test by Rob Fitzpatrick — finished July. Core idea: ask about past behavior, not opinions. Favorite line: people will lie to be nice.", "title": "book - The Mom Test"}'

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The skill requires a live API key and shows command examples that use it directly, but it does not warn users about secure credential handling. This can lead to accidental exposure through shell history, logs, screenshots, shared terminals, or copy-pasted examples in unsafe environments.

Content

No source excerpt is available for this finding.