ai-wrapper-product
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): Indirect prompt injection vulnerability surface detected in the provided implementation patterns.
- Ingestion points:
references/patterns.md(lines 35 and 110) contains code snippets showing user-controlled variables (userInput,input.purpose) being interpolated directly into prompt strings. - Boundary markers: Absent. The code examples do not demonstrate the use of delimiters (such as XML tags or triple quotes) to separate instructions from user-provided data.
- Capability inventory: The patterns demonstrate the ability to perform network operations to external AI providers using the
@anthropic-ai/sdk. - Sanitization: While basic length validation is present, there is no evidence of content sanitization or escaping to prevent the user input from overriding the system instructions.
Audit Metadata