memory-palace-architect
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of persistent knowledge schemas from user-provided input, creating a surface for indirect prompt injection where malicious instructions could be stored in domain descriptions and later influence the agent during retrieval.
- Ingestion points: User-provided strings for palace names and domains in the creation script (referenced in SKILL.md).
- Boundary markers: No delimiters or explicit instructions to ignore embedded content are defined in the architectural guidelines.
- Capability inventory: The skill uses local scripts to manage file-based knowledge structures.
- Sanitization: No sanitization or validation protocols for user-supplied descriptions are mentioned in the provided modules.
Audit Metadata