memory-audit
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for reading and classifying content from .claude/memory/MEMORY.md, which is a surface for indirect prompt injection if the memory file contains malicious instructions.
- Ingestion points: .claude/memory/MEMORY.md and knowledge/concepts/*.md (SKILL.md).
- Boundary markers: No explicit delimiters or boundary markers are defined for the content ingestion.
- Capability inventory: The skill uses Read, Write, Edit, Grep, and Glob tools to manage local files.
- Sanitization: No specific sanitization or filtering is mentioned; however, the risk is mitigated by a mandatory human-in-the-loop checkpoint ('Step 3 — PAUSE') which requires explicit user approval of the audit plan before any file modifications are performed.
Audit Metadata