workflow-integration-github

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

• Third-party content exposure detected (high risk: 0.78). Outsider-authored free text (GitHub PR review comment bodies) is fetched at runtime via gh in github_ops.fetch_pr_comments_data and then stored as raw_input={'body': body} by github_pr.py’s fetch_findings, which is later promoted into LLM-readable top-level fields during manage-findings ingest (validate_struct), creating indirect prompt-injection exposure.