The Agent Skills Directory

[SAFE]: The skill consists of markdown instructions and metadata used to guide the agent's reasoning. It does not include executable scripts or code blocks.
[SAFE]: No network operations, remote downloads, or external code execution patterns were detected. All model references are descriptive of third-party AI services and internal vendor integrations.
[SAFE]: There are no signs of data exposure, credential theft, or exfiltration. The skill does not access the file system or environment variables.
[SAFE]: The content does not contain obfuscation, malicious prompt injections, or persistence mechanisms. The 'expert advisor' persona is a standard instructional technique and does not attempt to override platform safety filters.

model-selector