The Agent Skills Directory

[PROMPT_INJECTION]: No prompt injection attempts were detected. The instructions focus on guiding the agent to provide idiomatic Elixir metaprogramming advice.
[DATA_EXFILTRATION]: No network operations, sensitive file access, or data exfiltration patterns were identified.
[EXTERNAL_DOWNLOADS]: All external links point to official Elixir documentation on hexdocs.pm, which is a well-known and trusted service.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or unsafe package installations were found. The skill teaches local compile-time code generation using Elixir's native quote and unquote mechanisms.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or sensitive environment variable accesses were detected.
[SAFE]: The skill is purely educational, providing best practices for language-specific features (macros and DSLs) and explicitly discouraging unsafe practices like Code.eval_string.

elixir-meta-programming