The Agent Skills Directory

[SAFE]: The skill provides a set of behavioral guidelines for software development, focusing on minimizing over-engineering and adhering to TDD. No malicious patterns, obfuscation, or unauthorized data access attempts were detected.- [COMMAND_EXECUTION]: The instructions permit the agent to use standard version control commands, specifically git revert and git restore, to manage repository state. These operations include explicit safety conditions, such as checking for uncommitted changes before restoring, which aligns with standard development practices.- [INDIRECT_PROMPT_INJECTION]: The skill involves updating documentation files like CLAUDE.md and AGENTS.md to reflect architectural trade-offs and rules. While these files influence agent behavior, the instructions define a strict and minimal update procedure that limits the potential for unauthorized instruction injection.

yagni