arabic-nlp
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [SAFE]: The skill provides structured documentation and best practices for integrating Arabic NLP tools such as CAMeL Tools, Farasa, and PyArabic.
- [COMMAND_EXECUTION]: Includes a JavaScript script (scripts/list-vendors.mjs) that reads a project-level JSON file (vendors.json) to provide a consistent list of supported vendors. The script uses standard file system access within the project directory.
- [PROMPT_INJECTION]: The instructions contain strong safety guardrails, requiring human approval for high-risk actions and explicitly forbidding the invention of credentials or undocumented API behaviors.
- [DATA_EXFILTRATION]: The skill emphasizes data privacy, specifically instructing the agent to avoid sending sensitive Arabic text to external APIs by default.
Audit Metadata