Funai-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the Agent to obtain the user's Token and "代为写入配置文件" (write it into config/.env), which requires the LLM to receive and output the secret value verbatim, creating an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). 在运行时工作流会读取并使用用户在“提交剧本/剧本内容(content)”处提供的自由文本（并通过 POST /project/step/next step=novel_input 将其提交给后端），因此外部作者可通过用户输入注入任意文本影响后续生成步骤。

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). scripts/api-client.sh fetches and parses https://neirong.funshion.net/skills/setup-skill.md at runtime (curl in fetch_setup_metadata_json) and many runtime helpers call require_setup_ready (so execution of create/next/modify actions is gated on that remote metadata which can set force_update/versions and thus directly control whether the skill proceeds)