The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions were found that attempt to override AI behavior or bypass safety filters. The content is strictly instructional and follows standard markdown patterns.
[Data Exposure & Exfiltration] (SAFE): The skill does not access sensitive files or hardcode credentials. Examples like you@example.com and aws ce get-cost-and-usage are used as placeholders for legitimate development purposes.
[Remote Code Execution] (SAFE): While the documentation describes how to install plugins from GitHub using claude plugin add, these are instructions for the human developer and not automated execution of untrusted remote code within the skill itself.
[Obfuscation] (SAFE): No Base64 encoding, zero-width characters, or other obfuscation techniques were identified in the provided file.
[Command Execution] (SAFE): Examples of bash commands (like tail, cp, and aws) are appropriate for the context of a development guide and do not represent a security risk.

plugin-development