Security audit

Personal Ideas

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent local idea-memory helper, but it can trigger broadly and automatically persist personal or business ideas without explicit per-save consent.

Review before installing if you discuss sensitive ideas, business plans, client work, or inventions. The skill is local and purpose-aligned, but you should only use it where automatic persistent idea capture is acceptable, and you should periodically inspect or clear knowledge/personal/ideas.md.

A.I.G

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

SkillSpector

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 (4)

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The skill description says it should activate when a user mentions very generic terms like "idea," "thought," "concept," "brainstorm," "project idea," or "what if." These phrases are common in ordinary conversation, so the skill may trigger outside the user’s intent and unexpectedly store or process personal content in the ideas knowledge file.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The activation description includes broad everyday phrases like idea, thought, concept, startup, project idea, and what if, which can cause the skill to trigger in many normal conversations outside the user's clear intent. In a personal-memory skill, over-triggering is more dangerous because it can lead to unnecessary reading of private notes and unprompted persistence of sensitive brainstorms.

Natural-Language Policy Violations

Medium

Confidence: 89% confidence
Finding: The skill title and operating instructions are written in Russian and specify tone and behavior in Russian without indicating that language choice is optional. This can violate language/locale policy when the user has not opted into Russian or been offered an alternative.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The skill instructs the agent to automatically store any new idea in knowledge/personal/ideas.md, but it does not require clear notice or consent about persistent retention. Because ideas may contain sensitive personal, business, or intellectual-property information, silent storage increases privacy risk and can capture data the user did not expect to be retained.

Static analysis

No suspicious patterns detected.