famou-experiment-manager

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs prompting the user for an API_KEY and inserting it verbatim into a command (python3 scripts/config.py write <YOUR_API_KEY>), which requires the LLM to handle/output the secret directly and creates exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly runs famou-ctl against an external API and parses JSON logs/results/status (see SKILL.md commands like "famou-ctl experiment logs/results/status" and the "解析 JSON 输出，展示实验 ID、状态等关键信息" / retry/delete logic), so untrusted remote experiment content can be ingested and directly influence follow-up actions such as deleting or resubmitting experiments.