memory-status
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
mi cortex statuscommand to retrieve diagnostics such as memory totals and ingestion freshness. It also mentionsmi doctorfor debugging. - [PROMPT_INJECTION]: The skill processes output from the
mitool, which constitutes an indirect prompt injection surface. - Ingestion points: Tool output from
mi cortex status(SKILL.md). - Boundary markers: None; there are no specific markers or instructions to treat tool output as untrusted data.
- Capability inventory: No high-privilege capabilities such as network exfiltration or file writing are present in the skill.
- Sanitization: None; output is processed directly by the agent without validation.
Audit Metadata