byted-sol-stability-grafana-metric-explorer

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

• Third-party content exposure detected (high risk: 0.75). At runtime, the required workflow reads outsider-authored free text from the provided --arch-model directory files (e.g., core-links.md, dependency-risk.md, observability-gaps.md) via load_arch_model() and passes their contents into the LLM context through normalize_inputs() → Pipeline.run_pipeline() (no LLM call is shown here, but the data is explicitly ingested as text for downstream generation).