managing-hubspot

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow calls HubSpot REST APIs (e.g., /automation/v4/flows, /crm/v3/pipelines/*, /marketing/v3/emails/statistics) and ingests the returned JSON fields into the agent context via jq-formatted output; these responses are outsider-authored data from a third-party service (HubSpot) that the operating user did not author.