The Agent Skills Directory

[SAFE]: The skill implements a standard Test-Driven Development workflow focused on local project maintenance and test automation. All identified behaviors are consistent with its primary purpose.
[COMMAND_EXECUTION]: The skill identifies and runs test commands (e.g., npm test, pytest, go test) based on the local environment's configuration files (e.g., package.json, pyproject.toml). This execution is limited to the project's own test runner.
[DATA_EXPOSURE]: The skill reads project metadata files and existing test code to align with the local project's standards. No access to sensitive files (e.g., .ssh, .env, .aws) or credential harvesting was detected.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user input via $ARGUMENTS and reads project files which could theoretically contain malicious instructions. However, the skill maintains a rigid operational loop (Red, Green, Refactor phases) with clear instructions that focus the agent on specific testing tasks, reducing the risk of indirect instruction obedience.

tdd