The Agent Skills Directory

[PROMPT_INJECTION]: The skill does not contain instructions that attempt to override agent behavior, bypass safety guidelines, or extract system prompts. The persona is strictly limited to that of a Go CLI engineer.
[DATA_EXFILTRATION]: There are no patterns indicating data exfiltration or unauthorized access to sensitive files. Network operations are not present in the code examples, and configuration management follows best practices by using environment variables and local config files.
[REMOTE_CODE_EXECUTION]: The skill does not perform or encourage the execution of untrusted remote code. It utilizes standard Go build and linting tools. Code examples focus on local development and compilation.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or secrets were found. The examples use placeholders for configuration values (e.g., local PostgreSQL DSNs) and correctly recommend managing secrets via environment variables or configuration files.
[COMMAND_EXECUTION]: Command execution is restricted to a set of allowed development tools (go, git, golangci-lint) as defined in the skill's frontmatter. No arbitrary shell execution or unsafe command construction was detected.
[SAFE]: The skill promotes secure development practices, including the use of signal.NotifyContext for graceful shutdowns, the separation of program output (stdout) from diagnostics (stderr), and the use of built-in argument validation.

golang-cli