code-rot-cleaner

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

• Third-party content exposure detected (high risk: 0.78). The required workflow reads repository files (which may include outsider-authored source text like public repo contents or downloaded attachments) via scripts/analyze_code_rot.py using safe_text() → path.read_text(...), then writes that extracted text into JSON artifacts (analysis.json) that are later loaded and rendered into the report (scripts/generate_report.py), making it LLM-readable free text for the agent.