prompting-company

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires fetching live documentation and machine-readable OpenAPI specs at runtime (e.g., https://docs.promptingcompany.com/api-reference/openapi.json and https://app.promptingco.com/api/v1/openapi.json), and instructs the agent to use those fetched pages as authoritative input to determine endpoints, scopes, and commands—meaning external content will directly control prompts/behavior.