playbook-case-study-page

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

• Third-party content exposure detected (low risk: 0.10). Outsider-authored free text is ingested only when the workflow fetches and scrapes a prospect domain’s own public case-study/customer-stories page (HTTP probes -> scrape-website -> page text -> model picker with strict quote gates), and there is no queue/feed/email/chat style submission path for an outsider to inject arbitrary text.

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

• Potentially malicious external URL detected (high risk: 1.00). The workflow fetches prospect web pages at runtime and injects the fetched page text into the model prompt (e.g., GET https://{{NormalizedDomain}}/customers and the constructed URL pattern https://%s%s), so remote web content like https://ramp.com/customers directly controls the agent's outputs.