ClawAudit verdict
model-platform-guide
A reference and documentation skill for Chinese LLM API platforms (Kimi, DashScope, SiliconFlow, DeepSeek, etc.) providing API documentation links, usage examples, and comparison tables — purely a knowledge/reference skill with no code execution.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (3)
Possible hardcoded credential
references/moonshot.md · code · api_key="YOUR_MOONSHOT_API_KEY
Python os.getenv — reads environment variable
references/aliyun-bailian.md · code · os.getenv(
Accesses sensitive environment variables
references/aliyun-bailian.md · code · process.env.DASHSCOPE_API_KEY
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.