langsmith-fetch

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

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text can enter the LLM context via LangSmith Studio trace/thread content fetched at runtime by langsmith-fetch traces ... / langsmith-fetch trace <id> ... (trace steps, tool inputs/outputs, error messages, and metadata may include user messages/comments from other parties), which the agent then analyzes and reports.

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

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