arkcli-auth

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to accept a user-provided base64 SSO authorization code and run a CLI command embedding it verbatim (arkcli auth login --no-browser --code <授权码>), which requires the LLM to handle/output a secret value directly.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires using the Read tool at runtime to load ../arkcli-shared/SKILL.md, and that fetched file is a required dependency that directly controls authentication gating and agent prompt/decision behavior (../arkcli-shared/SKILL.md).