investigation-entrypoint

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow ingests PagerDuty/incident-context input and immediately uses it to trigger gcp-architecture-discovery and subsequent logging/metrics investigation, meaning outsider-authored incident/event text delivered via the PagerDuty→integration path could be read by the subagents/tools before any filtering for a specific item.