The Agent Skills Directory

[SAFE]: The skill is entirely educational and referential. It contains markdown documentation and code snippets illustrating standard mobile development patterns such as MVVM, Clean Architecture, and reactive UI development using modern frameworks.
[CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The skill explicitly identifies 'Hardcoded credentials' as a common pitfall and recommends using secure storage solutions like iOS Keychain and Android KeyStore, as well as environment variables for configuration.
[EXTERNAL_DOWNLOADS]: The skill references numerous external resources, including official documentation for React Native, Flutter, Apple Developer, Material Design, and Firebase. All identified URLs point to well-known, trusted technology services and organizations (e.g., Google, Apple, Meta, OWASP).
[DATA_EXFILTRATION]: No evidence of data exfiltration was found. Network-related code snippets (e.g., using fetch or URLSession) are purely illustrative examples of how developers should interact with their own backend APIs.
[PROMPT_INJECTION]: The SKILL.md file and reference documents do not contain instructions aimed at overriding agent behavior or bypassing safety guidelines. The instructional language is focused on mobile development best practices.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not use the dynamic context injection syntax (!command) to execute shell commands at load time.
[OBFUSCATION]: No obfuscated content, multi-layer encoding, homoglyph attacks, or hidden text patterns were detected in the skill files.

mobile-development