recipe-verify-ascii-diagrams

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

• Third-party content exposure detected (high risk: 0.70). The required workflow runs pm-documents:ref-ascii-diagrams:ascii_diagrams check/fix --path {file} over candidate .md/.adoc files discovered from marketplace/bundles and doc, so the LLM context can ingest outsider-authored free text from those repository files at runtime (public/third-party content surface), creating indirect prompt-injection risk.