Back to skill

Security audit

Music Discovery

Security checks for vulnerabilities and agentic risk

Overview

This skill presents itself as a music recommendation tool, but its bundled command mainly records user inputs locally and asks for unused unpinned dependencies.

Review before installing. The skill does not appear to exfiltrate data or perform destructive actions, but it is over-described compared with what it actually does, stores your music-related inputs locally, and asks you to install unnecessary unpinned Python packages. Avoid entering secrets or account tokens as arguments, and prefer removing or pinning the unused dependencies before use.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:18
Finding

Unpinned Third-Party Dependencies Installed Without Integrity Verification

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (2)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The module docstring presents the tool as a 'Music discovery' CLI with recommend/playlist/mood flows, and the function docstrings similarly describe substantive music operations. In practice, each command just stores the provided arguments in a local JSON file and returns a generic completion message plus static reference URLs, which contradicts the documented intent of actually discovering or generating music-related results.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

User-supplied command arguments are written verbatim to a JSON file on disk without any disclosure, minimization, or redaction. In this skill context, users may pass personal preferences, listening habits, account-related identifiers, or other sensitive text as CLI arguments, creating a local privacy leak and a durable record that could be exposed to other local users, backups, or later processes.

Content

No source excerpt is available for this finding.