The Agent Skills Directory

[SAFE]: The skill manages user preferences by reading and writing to data/user-settings.json. This usage of the Read and Write tools is limited to managing internal application state and does not expose sensitive system files, environment variables, or credentials.
[SAFE]: Automated language detection is implemented as a simple character-ratio calculation performed by the agent. This process does not involve external API calls, remote code execution, or the processing of untrusted executable content.
[SAFE]: No malicious patterns such as prompt injection, obfuscation, or persistence mechanisms were detected in the skill instructions, examples, or data schema.
[SAFE]: The skill does not declare or use any external Python or Node.js dependencies, nor does it attempt to download or execute code from remote sources.

language