bigdata-skill

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

• Third-party content exposure detected (high risk: 0.85). At runtime the skill ingests outsider-authored free text only via the SDK search path (AnnotatedSearcher.search_entity → bd.search.new(...).run(...)), which returns news/document chunk text authored by third parties; those chunk texts are then flattened into the LLM context as docs[*].chunks[*].text (and related sentiment/entity spans).

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)