The Agent Skills Directory

[PROMPT_INJECTION]: No prompt injection or behavior override attempts were detected. The instructions are focused on providing idiomatic Kotlin development guidance.
[DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were identified. Code examples use standard placeholders for API endpoints and database connections.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or secrets were found. References to authentication (JWT) and publishing (GitHub Tokens) correctly use environment variables and configuration properties as per best practices.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as piping network downloads to a shell, were found. All dependencies mentioned are standard Kotlin/JVM libraries.
[COMMAND_EXECUTION]: No dangerous shell command execution was detected. The skill does not use dynamic context injection (!command) patterns.
[OBFUSCATION]: No obfuscated code, hidden characters, or encoded payloads were found in any of the analyzed files.
[PERSISTENCE_MECHANISMS]: The skill does not attempt to modify system startup files, shell profiles, or scheduled tasks.
[PRIVILEGE_ESCALATION]: No commands for acquiring administrative privileges (sudo, chmod 777, etc.) were identified.
[DYNAMIC_EXECUTION]: No runtime code generation or unsafe deserialization patterns were detected. The skill focuses on static code templates.

kotlin-specialist