The Agent Skills Directory

[NO_CODE]: The skill consists entirely of Markdown and metadata instructions for the AI agent. It does not include any executable scripts, binaries, or configuration files that perform actions on the host system. All logic is expressed through natural language prompts.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external code provided to the agent for simplification. This creates a surface where the processed code could theoretically contain instructions intended to influence the agent. However, the skill includes robust procedural safeguards, such as the requirement to 'Preserve Behavior Exactly', 'Understand Before Touching', and verify changes against existing tests, which mitigates the risk of the agent following instructions embedded in the data it is refactoring.
[EXTERNAL_DOWNLOADS]: The README.md references a source URL from a well-known developer (Addy Osmani) on GitHub. This is a neutral documentation reference to the skill's provenance and does not involve automated downloads or execution during the skill's operation.

simplify