agent-asset-studio

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow routes by reading and then loading the selected workflow module(s) and resources from the repo’s workflows/INDEX.csv and corresponding workflows/<id> files, so any outsider-authored Markdown placed into those paths (e.g., from issues/PRs/wiki updates) would be ingested as free text into the LLM context via the loaded workflow files.