The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and metadata focused on project architecture. No code, scripts, or binaries are included in the package.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input such as app ideas or project visions. However, the skill lacks the capability to execute commands, perform network operations, or access files, meaning there is no exploitable surface for an injection attack beyond influencing the agent's textual output.
[NO_CODE]: There are no executable files, package dependencies, or shell commands present. The skill operates purely within the context of text-based planning.

mode-vision