memory-kit
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from a shared MEMORY.md file without isolation, creating an injection surface. Ingestion points: SKILL.md instruction 1 triggers reading and summarizing the file at session start. Boundary markers: Delimiters or 'ignore instructions' warnings are missing. Capability inventory: The skill possesses Read, Write, Edit, and Bash permissions. Sanitization: No validation or filtering is applied to the memory file content.
- [DYNAMIC_CONTEXT_INJECTION]: The skill uses load-time shell placeholders to report file status. Evidence: Shell checks for MEMORY.md and tasks/current-task.md in SKILL.md. Analysis: These operations are restricted to benign metadata checks and do not incorporate untrusted arguments or perform unauthorized actions.
Audit Metadata