memory-remember
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the vendor's 'mi' CLI tool ('mi cortex remember') to save information to a persistent memory store. This is the primary and intended core functionality of the skill.
- [PROMPT_INJECTION]: The skill provides an interface to save potentially untrusted text into persistent memory. 1. Ingestion points: Data provided to the argument in the CLI command. 2. Boundary markers: The command template uses double-quote delimiters for the text input. 3. Capability inventory: The skill uses the 'mi' CLI tool to write to the memory system. 4. Sanitization: No explicit input sanitization or validation of the text content is described.
Audit Metadata