render-reading-diff

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to "save the complete diff" and to "save supplied diff verbatim" and to render a self-contained HTML review including the original diff, which would require outputting any secrets present in the diff verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow Render Reading Diff ingests outsider-authored free text by accepting arbitrary “Supplied diff” content and directly rendering it into an HTML report via scripts/lib/report.sh → awk/render-diff.awk, without requiring the diff to be preselected from a trusted list.