The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run standard Elixir build and quality tools including mix test, mix format, mix credo, and mix dialyzer. These commands are standard for the intended TDD workflow and do not involve unauthorized shell operations.
[DATA_EXFILTRATION]: No network-enabled commands (e.g., curl, wget) or access to sensitive system directories (e.g., .ssh, .aws, .env) are present in the instructions.
[PROMPT_INJECTION]: The instructions follow a structured workflow and do not contain language intended to bypass safety filters or override system-level instructions.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or private secrets were found in the skill metadata or instructional examples.
[REMOTE_CODE_EXECUTION]: The skill does not download or execute scripts from external or untrusted URLs. All operations are performed on local project files within the Elixir environment.

tdd