iterate-ml-experiment

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow can ingest outsider-authored free text via the iterate-from-user sourcing menu: the user may paste an article URL / GitHub issue / spec path / free-text idea, which the skill then reads and synthesizes into the agent’s LLM context (indirect prompt-injection risk from outsider-provided content).