echo-ingest-workflow-prd

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s workflow is described as ingesting “messy source material” via the referenced intake step ($echo-ingest-source-intake), which at runtime would commonly include outsider-authored free text (e.g., issue/PR/discussion body text or forwarded attachments) that is then extracted and fed into the LLM context for PRD generation.