The Agent Skills Directory

[SAFE]: The skill provides standard architectural patterns for implementing JWT authentication and authorization in Go Gin applications. It follows security best practices such as generic error messages for authentication failures, password hashing using bcrypt, and explicit validation of JWT signing methods to prevent algorithm switching attacks.
[EXTERNAL_DOWNLOADS]: It references well-known and trusted Go packages: github.com/golang-jwt/jwt/v5 and golang.org/x/crypto. These are industry-standard libraries for handling tokens and cryptography in the Go ecosystem.
[CREDENTIALS_UNSAFE]: No hardcoded secrets or sensitive credentials are present in the code. The provided examples correctly demonstrate loading secrets and configuration from environment variables (e.g., using os.Getenv).
[PROMPT_INJECTION]: No instructions aimed at overriding agent behavior or bypassing safety guidelines were found in the prompts or documentation.
[COMMAND_EXECUTION]: The skill does not perform any unexpected system command execution or shell spawning. The server setup and signal handling are standard for Go web applications.

gin-auth