a1-yandex-kit

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides local Node.js scripts (e.g., scripts/search_docs.mjs) for searching the Yandex KIT API specification. These scripts are self-contained, using only built-in Node.js modules, and do not perform network operations.
  • [DATA_EXFILTRATION]: No network-enabled code or unauthorized data transfer patterns were detected. The skill is designed for offline use of the API specification.
  • [PROMPT_INJECTION]: The skill includes merchant communication guidelines to standardise agent responses, but does not contain any adversarial instructions aimed at bypassing model safety or overriding core behavior.
  • [SAFE]: The skill demonstrates safe security practices, including clear guidance on API token management and warnings against production writes during exploration.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 01:27 PM
Security Audit — agent-trust-hub — a1-yandex-kit