Back to skill

Security audit

Voice Notes

Security checks for vulnerabilities and agentic risk

Overview

This skill organizes voice transcripts into local notes and clearly limits itself to local storage without external access or hidden execution.

Before installing, understand that your transcribed voice messages will be stored as local Markdown files under ~/voice-notes/, including raw transcript text. Avoid using it for highly sensitive recordings unless you are comfortable with that local retention and manage file access appropriately.

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
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • 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)

Memory Manipulation

High
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

Content

Scanner excerpt · processing.md (reported line 37)May include surrounding context.

md
| Input | Output |
|-------|--------|
| "um, so like..." | Clear statement |
| Repeated points | Single consolidated point |
| Tangents | Separate linked notes |
| Stream of thought | Logical sections |

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly instructs saving raw transcripts before any processing, which can persist highly sensitive speech-derived content such as personal data, secrets, or confidential discussions. Because there is no mention of user consent, minimization, retention limits, or access controls, this creates a real privacy and data-exposure risk if the notes store is accessed improperly or retained longer than expected.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.