ClawHub
Back to skill

Security audit

Ai Sidekick

Security checks for vulnerabilities and agentic risk

Overview

This skill is not malware, but it broadly enables automatic persistent memory and backups of personal/project details without clear consent or retention controls.

Install only if you intentionally want this skill to maintain persistent local memory about you and your projects. Before using it, decide what may be written to USER.md, MEMORY.md, memory logs, and backups; avoid storing secret values or secret-adjacent locations; and periodically inspect or delete the retained memory files and backup archives.

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 (4)

Vague Triggers

High
Confidence
95% confidence
Finding
The trigger description is extremely broad and includes generic requests like configuring AI behavior, training AI, security settings, and memory management. This can cause the skill to activate in contexts where the user did not clearly intend persistent configuration or memory-related actions, increasing the chance of inappropriate data handling or instruction takeover.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs storing personal preferences, identity details, project context, and conversation-derived information in persistent files without a prominent retention notice or consent workflow. This creates a privacy risk because users may disclose information in ordinary conversation without realizing it will be retained across sessions.

Ssd 3

Medium
Confidence
97% confidence
Finding
The long-term memory guidance explicitly stores user identity, preferences, project decisions, and even operational details such as API key locations and certificate paths. Persisting sensitive metadata in natural-language memory files increases the chance of later accidental disclosure, oversharing in prompts, or exposure through filesystem compromise.

Ssd 3

Medium
Confidence
96% confidence
Finding
The backup script archives MEMORY.md and the memory directory into compressed backups under a predictable local path without mentioning encryption, access control, or retention policy. Backing up sensitive conversational and operational data expands the attack surface and increases the blast radius of any local compromise or accidental sharing.

Static analysis

No suspicious patterns detected.