llmlb-cli-usage
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill allows the ingestion of untrusted data from API schemas via the openapi command. • Ingestion points: llmlb assistant openapi. • Boundary markers: None mentioned. • Capability inventory: Bash and network access. • Sanitization: Skill documentation claims internal tool filtering, which is unverified.
- [COMMAND_EXECUTION] (SAFE): Bash tool access is necessary for the CLI utility.
- [DATA_EXPOSURE] (SAFE): Authentication variables are referenced for configuration only; no credentials are leaked.
Audit Metadata