The Agent Skills Directory

[SAFE]: The skill is a documentation and methodology resource for prompt optimization, focusing on transforming negative framing into positive instructions based on official guidelines.
[NO_CODE]: The skill consists entirely of Markdown and YAML configuration files. No executable scripts (Python, JavaScript, Shell) are included in the skill package.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process and refactor prompt configuration files, which represent an indirect injection attack surface. The methodology emphasizes structural boundaries (XML tags, headers) and multishot examples, which are standard safety and performance techniques for LLMs.
[EXTERNAL_DOWNLOADS]: The skill references documentation URLs from trusted and well-known domains, specifically docs.claude.com and platform.claude.com, to ensure instructions are based on official Anthropic sources.
[COMMAND_EXECUTION]: Examples within the skill recommend using the Bash tool for environment management with uv run, which is a legitimate and recommended practice for managing Python project dependencies.

prompt-optimization