google-form-callback

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

• Third-party content exposure detected (medium risk: 0.30). In scripts/google-local-api-client.mjs → scripts/extract-callback-candidates.mjs → scripts/process-callback-candidates.mjs, the workflow ingests Google Form response text (respondent-authored) and compiles it into the provider goal/summary fields for CALL-E based on that untrusted content.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches Google Form metadata/responses at runtime (e.g., GET https://forms.googleapis.com/v1/forms/{formId} and https://forms.googleapis.com/v1/forms/{formId}/responses) and also supports calling an Apps Script Web App (https://script.google.com/macros/s/.../exec), and those fetched response fields are compiled into the provider "goal" that controls CALL-E prompts, so external data directly controls runtime prompts.