Back to skill

Security audit

Drums

Security checks for vulnerabilities and agentic risk

Overview

This drum-practice skill stores local progress notes for its stated tracking purpose, with no evidence of hidden code, exfiltration, or destructive behavior.

Before installing, understand that the skill may create a ~/drums/ folder and keep local notes about practice sessions, songs, goals, and tempos. This looks low risk, but users who do not want local habit tracking should decline logging prompts or remove the folder.

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 creation of a local workspace in the user's home directory on first interaction without explicit consent or notice. Even though `~/drums/` is not inherently sensitive, writing to the filesystem without confirmation violates least surprise and can create unwanted local artifacts or normalize unauthorized file operations by agent skills.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The logging trigger is broad enough to capture routine conversation about practicing or playing, which can cause the agent to prompt for logging in situations where the user did not clearly intend to record activity. While this is not overtly malicious, it creates an avoidable privacy and UX risk by encouraging unnecessary data collection and intrusive prompts.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill instructs logging practice data to `~/drums/` without an explicit warning that user activity data will be stored locally. While the content appears low sensitivity, persistent storage of habits, goals, and progress still affects privacy expectations and should not occur silently.

Static analysis

No suspicious patterns detected.