memory
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a vulnerability to indirect prompt injection by processing content from external memory files that may contain malicious instructions.\n
- Ingestion points: The skill ingests data from
USER.md,MEMORY.md, and various files within thememory/andturns/directories (SKILL.md).\n - Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore potentially malicious embedded instructions within retrieved data (SKILL.md).\n
- Capability inventory: The skill utilizes filesystem tools,
memory_search,memory_get,memory_save, andmemory_deleteto manage and interact with memory files (SKILL.md).\n - Sanitization: The skill lacks explicit instructions for sanitizing, validating, or filtering content retrieved from the file system before interpolation into the agent's context (SKILL.md).
Audit Metadata