The Agent Skills Directory

[SAFE]: The skill provides structured guidance for development tasks without any detected malicious behaviors. It adheres to best practices for workspace isolation and dependency management via git worktrees and lockfiles.- [COMMAND_EXECUTION]: The skill instructions involve executing standard local development commands (git, package managers, and test runners) to prepare workspaces and run tests. All operations are restricted to the local filesystem and intended development context.- [INDIRECT_PROMPT_INJECTION]: The skill ingests external task descriptions and specifications. While this is a theoretical attack surface, the risk is minimal as the skill instructions focus on local file operations and testing within a controlled development environment, following a structured TDD and review process.

implement