Back to skill

Security audit

Bass

Security checks for vulnerabilities and agentic risk

Overview

This bass practice skill is a simple Markdown-only helper with local progress tracking, but users should know it may create and use a local ~/bass/ folder.

Install only if you are comfortable with the agent creating a ~/bass/ folder and saving bass practice notes there. Ask the agent not to log practice, or to use a different location, if you do not want local progress tracking.

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
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 (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill directs the agent to create a directory in the user's home folder on first interaction without first informing the user or requesting consent. Even though the data appears low sensitivity and the skill is music-focused, silent filesystem writes create an unnecessary privacy and trust risk and can normalize unexpected local persistence.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill proposes logging practice progress to files under `~/bass/` but does not provide an upfront warning that user activity will be stored locally. While the content is not inherently sensitive, progress logs can still reveal habits, schedules, and personal notes, so undisclosed persistence is a legitimate privacy issue.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The logging triggers are intentionally broad and could cause the agent to prompt for logging during ordinary conversation, leading to unnecessary collection or retention of user activity data. In a progress-tracking skill this is contextually related behavior, but the lack of explicit boundaries, consent checks, or cooldown rules still creates a real over-collection and annoyance risk.

Static analysis

No suspicious patterns detected.