Validating OpenAI API Implementations

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs the agent to fetch and parse the OpenAPI spec from an external URL (https://app.stainless.com/api/spec/documented/openai/openapi.documented.yml, including a curl example in "Recommended Approach"), so untrusted third‑party YAML content is ingested and used to drive validation and decision-making, enabling indirect prompt injection risk.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs fetching and using the OpenAPI spec at runtime via curl (https://app.stainless.com/api/spec/documented/openai/openapi.documented.yml), which would directly control validation logic and agent behavior based on remote content.