arabic-nlp
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from external sources. \n
- Ingestion points: Reading vendor-specific facts from
vendors/*.mdand source data fromsources.yml. \n - Boundary markers: No explicit delimiters are mentioned in the instructions. \n
- Capability inventory: Used for tool identification and implementation planning. \n
- Sanitization: Requires source-backed answers and explicit labeling of unknowns. \n- [COMMAND_EXECUTION]: The workflow references the execution of a local script
scripts/list-vendors.mjsto generate a vendor list.
Audit Metadata