Security audit
Security checks for vulnerabilities and agentic risk
This is a storytelling-writing helper that stores optional local narrative notes and does not include code execution, network behavior, or hidden privileged actions.
Before installing, understand that the skill may keep local storytelling notes such as audience assumptions, reusable anecdotes, voice preferences, and edit history under ~/storytelling/. This is useful for continuity, but do not let it save sensitive personal details, confidential business facts, credentials, or private source material unless you deliberately want that retained locally.
The 'When to Use' section describes very general situations like needing to 'explain, persuade, or teach through narrative,' which overlap with many ordinary writing tasks. It does not provide explicit trigger phrases, constraints, or negative examples to clarify when this skill should activate versus when other writing-related skills should be used.
The activation condition is overly broad because 'always on for storytelling requests' lacks clear boundaries for what counts as a storytelling request. This can cause the skill to activate in unrelated contexts, leading to inappropriate behavior shaping, unexpected data handling, or silent influence over outputs beyond the user's intent.
The instruction to store user context so future sessions start with the same constraints creates persistent retention behavior without a clear user-facing notice or consent step. In a skill that captures audience profiles, beliefs, objections, and preferences, this can result in unexpected retention of personal or sensitive profiling data across sessions.
No suspicious patterns detected.
