instructor

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • PROMPT_INJECTION (LOW): Indirect Prompt Injection Surface.
  • Ingestion points: The variable text in references/examples.md (e.g., f"Extract: {text}") is designed to ingest untrusted external data.
  • Boundary markers: Absent; the code snippets show direct string interpolation without using delimiters (like XML tags or triple quotes) or 'ignore embedded instructions' warnings.
  • Capability inventory: The snippets demonstrate using the LLM to process this data and populate Pydantic models via client.messages.create.
  • Sanitization: Absent; there is no evidence of input validation or escaping before the data is passed to the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:26 PM