tldr
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: This skill is entirely comprised of markdown instructions and does not contain any executable scripts, binary files, or external network dependencies.
- [NO_CODE]: There are no code files or dependencies shipped with this skill.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface through its core data processing flow:
- Ingestion points: The skill processes the full conversation history from the agent's context to generate summaries.
- Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore embedded commands within the conversation history.
- Capability inventory: The skill utilizes file-writing capabilities to create notes in vault subdirectories and append updates to the 'memory.md' file.
- Sanitization: No sanitization or data validation is performed on the summarized content before it is saved to the filesystem.
Audit Metadata