agent-tool-builder
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a comprehensive educational resource for developing agent tools. It focuses on schema design, error handling, and the use of official SDKs from reputable providers.
- [INDIRECT_PROMPT_INJECTION]: The skill discusses tools that ingest untrusted data from external sources such as weather APIs and stock tickers. While this represents a potential attack surface, the skill explicitly provides mitigation strategies.
- Ingestion points: External data sources like
weather_api.fetchandget_stock_price(SKILL.md). - Boundary markers: Explicitly recommends validation gates and error handling to prevent agents from being poisoned by silent failures.
- Capability inventory: Uses standard SDK functions for tool calling and result processing.
- Sanitization: Strictly enforces SQL parameterization and input validation before execution.
Audit Metadata