qinghu-media-tools

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs the agent to ask the user for an API key token and then embed it directly into a CLI command (qhkit config set --token <密钥>), which exposes the secret value in the LLM context and execution logs.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the agent to install and execute a third-party npm package (@iqinghu/qhkit) and download Node.js binaries from nodejs.org during runtime setup, representing a common/reputable unofficial/third-party operational dependency.