cn-llm
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill makes network requests to
api.aisa.oneto facilitate its primary function as an LLM gateway. This endpoint is owned by the skill vendor and is used to proxy requests to supported Chinese AI models. - [INDIRECT_PROMPT_INJECTION]: The skill acts as a conduit for untrusted data, creating an attack surface for indirect prompt injection.
- Ingestion points: The
scripts/cn_llm_client.pyscript accepts user messages, system prompts, and message arrays as input through CLI arguments and class methods. - Boundary markers: The skill does not utilize delimiters or specific instructions to mitigate the risk of the model obeying instructions embedded in the user content.
- Capability inventory: The skill is capable of performing network requests to the AIsa API to transmit prompts.
- Sanitization: User-provided text is included in the API request payload without sanitization or validation, which is standard behavior for an LLM client.
Audit Metadata