Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only knowledge search skill whose broad triggers may activate more often than intended, but its behavior is disclosed and purpose-aligned.
Before installing, understand that this skill may be invoked by broad words like search or recall. Use it when you want memory-palace lookup behavior, and check any referenced palace_manager.py implementation separately because it is not bundled in this artifact.
Using a trigger/command namespace centered on 'search' creates a collision risk with common built-in or platform search behavior. If this skill shadows a native search command, users may invoke the skill unintentionally, allowing it to alter retrieval behavior or inject memory-palace context where a normal search was expected.
The trigger list is broad and includes generic terms like retrieval, search, indexing, recall, and a long natural-language phrase that could match many ordinary user requests. This can cause unintended activation and context hijacking, especially because the skill is designed to influence how information is searched and surfaced during normal workflow and PR review tasks.
No suspicious patterns detected.
