moshi-best-practices

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt explicitly shows passing a pairing token on the command line (e.g., moshi-hook pair --token <YOUR_TOKEN>) and discusses writing host secrets to files, which instructs embedding secret values verbatim in commands or configs and thus creates an exfiltration risk.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs installing moshi-hook via the Homebrew tap (brew tap rjyo/moshi / https://github.com/rjyo/homebrew-moshi) which will be fetched at runtime and installs/executes remote code that the skill relies on for agent hooks.