search

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow is executing Elasticsearch search()/suggest() calls and then ingesting the returned _source/highlight/suggest text into the agent context; those texts originate from the Elasticsearch index (i.e., data authored/controlled by parties other than the operating user), which is outsider-authored free text read by the LLM via the search response parsing.