techdebt

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

• Third-party content exposure detected (high risk: 0.75). The required workflow runs scripts/scan.sh, which executes grep -rn ... "$SCAN_PATH" over the user-supplied path and ingests matching outsider-authored free text (e.g., TODO/FIXME/HACK comments) from files in that directory into the agent’s LLM context via the scan’s stdout report.