The Agent Skills Directory

[SAFE]: The skill is a set of documentation and procedural guidelines for developers or agents working on refactoring a specific repository. It contains no executable code, network operations to untrusted domains, or sensitive data access patterns.
[COMMAND_EXECUTION]: The skill mentions using standard development tools such as rg (ripgrep), git, and pnpm. These are used for searching the codebase and performing repository maintenance (linting, documentation syncing) within the context of the developer's environment, which is the intended purpose of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill involves reading and processing repository content (code, tests, and documentation). While this presents a surface for indirect prompt injection if the repository contains malicious instructions, the skill's outputs are limited to creating markdown-based Agent Notes and inline code comments, posing minimal risk to the agent's execution environment.
[EXTERNAL_DOWNLOADS]: The skill provides guidance on evaluating and introducing NPM packages or Node.js built-ins as a means of simplification. It emphasizes following a specific internal dependency policy and checking for package health (maintenance, adoption), which aligns with secure development practices.

dsh-find-simplifications