kimi-result-handling

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly requires quoting kimi's output verbatim (no paraphrase/compression), which forces the model to reproduce any secret values present in kimi's response and thus enables secret exfiltration.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow for this skill is to render the companion-provided response string from kimi-companion.mjs (i.e., outsider-authored free text from Kimi) into the agent’s LLM context; this is not user-authored and is treated as untrusted content for presentation.