nat-tools-and-functions

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a reference for extending the NeMo Agent Toolkit (NAT), an official framework from a trusted vendor. The implementation patterns (subclassing, decorators) are standard for Python library extensions.
  • [EXTERNAL_DOWNLOADS]: The documentation references well-known and reputable third-party libraries including langchain-core, pydantic, fastapi, and httpx. These are industry-standard dependencies for building agentic and web applications.
  • [DATA_EXFILTRATION]: Example configurations for database connections and API workers use safe placeholders, such as localhost and 0.0.0.0. No hardcoded credentials, sensitive secrets, or unauthorized network exfiltration patterns were found.
  • [PROMPT_INJECTION]: The skill demonstrates tools that ingest structured data, such as SQL queries. While these define ingestion surfaces for potential indirect prompt injection, the examples are presented as developer templates rather than active or exploitable implementations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 11:05 AM
Security Audit — agent-trust-hub — nat-tools-and-functions