api-integration
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill involves analyzing external API documentation and OpenAPI specifications, which are potential vectors for untrusted data ingestion.
- Ingestion points: External API documentation URLs or user-provided documentation text mentioned in the workflow (SKILL.md).
- Boundary markers: None identified in the instructional content.
- Capability inventory: Network operation capabilities via HTTP clients like
requestsandaxios(SKILL.md). - Sanitization: The skill recommends validating status codes and parsing JSON structures, which provides a layer of protection against unexpected response content.
- [EXTERNAL_DOWNLOADS]: The skill provides examples of interacting with well-known and trusted services, including Stripe and GitHub.
- [CREDENTIALS_SAFE]: The skill demonstrates safe practices for secret management by retrieving API keys and tokens from environment variables rather than hardcoding them.
Audit Metadata