The Agent Skills Directory

[NO_CODE]: The skill consists entirely of Markdown instructions and design principles. It does not include any scripts, binaries, or automated command execution steps.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided architectural context and design constraints to generate interface proposals and depth analyses. While it ingests external data without explicit boundary markers, the scope is limited to reasoning and documentation, posing minimal security risk.
[SAFE]: All processes described, including the spawning of sub-agents for parallel design exploration, are standard collaborative patterns for architectural reasoning and do not involve unauthorized system access or data exfiltration.

codebase-design