redis
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the redis-cli binary to perform database operations such as GET, SET, and KEYS.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data from an external database.
- Ingestion points: Data retrieved via redis-cli GET and redis-cli KEYS commands.
- Boundary markers: No delimiters or instructions to ignore embedded commands are present in the prompt templates.
- Capability inventory: Shell command execution via redis-cli.
- Sanitization: No evidence of validation or sanitization of retrieved database content.
Audit Metadata