The Agent Skills Directory

[SAFE]: The skill provides comprehensive instructions for a standard software development lifecycle, focusing on TDD (Red-Green-Refactor), version control best practices, and project status tracking.
[SAFE]: The 'Security Audit' section explicitly instructs the agent to verify that no secrets are present in code, input validation is implemented, and dependencies are free of vulnerabilities.
[COMMAND_EXECUTION]: The instructions include usage of standard local development tools such as git, pytest, ruff, and mypy. These commands are appropriate for the skill's primary purpose of managing a development workflow and do not involve privilege escalation or unauthorized access.
[INDIRECT_PROMPT_INJECTION]: The skill processes project-local files such as plan.md and spec.md to track progress. While these represent an ingestion surface for untrusted data, the skill includes human-in-the-loop verification steps (e.g., 'WAIT for User Approval') before completing major phases, which mitigates risk.

workflow-patterns