The Agent Skills Directory

[SAFE]: The skill consists of pure instructional content designed to guide an AI agent through a structured dialogue process. No malicious patterns, command executions, or network requests were identified.
[NO_CODE]: The skill contains no scripts, binaries, or configuration directives that trigger tool execution or system modifications.
[INDIRECT_PROMPT_INJECTION]: While the skill process involves analyzing project context (files and config), it establishes strong human-in-the-loop controls by requiring explicit user approval for every design section before proceeding to implementation, mitigating the risk of autonomous obedience to embedded instructions in data.

brainstorming