effect-ai-provider
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill demonstrates patterns for building AI applications that are susceptible to indirect prompt injection by interpolating untrusted data into language model prompts.
- Ingestion points: External data enters the system through the
productandmessageparameters in theAiWriterservice examples inSKILL.md. - Boundary markers: The code snippets do not include delimiters (such as XML tags or triple backticks) or specific instructions to the model to treat interpolated content as data rather than instructions.
- Capability inventory: The skill enables text generation, streaming, and stateful chat management using various AI providers (Anthropic, OpenAI, OpenRouter).
- Sanitization: No input validation, escaping, or filtering is shown for the user-provided variables before they are used in prompt construction.
Audit Metadata