The Agent Skills Directory

[Prompt Injection] (SAFE): The skill uses strong instructional language to guide the agent in adhering to coding principles (SRP). While it includes a 'Pressure Resistance Protocol' to help the agent push back against poor architectural requests, these instructions are scoped to coding best practices and do not attempt to bypass safety guidelines or ignore the agent's core system prompt.
[Data Exposure & Exfiltration] (SAFE): No sensitive data, hardcoded credentials, or unauthorized network communication patterns were detected. The TypeScript files define clean interfaces for authentication and storage without exposing implementation details or secrets.
[Remote Code Execution] (SAFE): There are no commands for downloading external scripts or executing arbitrary code. The project lacks package manifest files (e.g., package.json) and does not utilize dangerous functions like eval() or subprocess calls.
[Privilege Escalation] (SAFE): The skill does not attempt to gain administrative access or modify system permissions.
[Obfuscation] (SAFE): All files are written in clear, human-readable Markdown and TypeScript. No hidden characters, Base64 encoding, or homoglyph attacks were found.

single-responsibility-principle