What problem does it solve?
Building and maintaining ChatGPT Apps SDK projects is complex, slow, and error-prone without a guided, docs-aligned scaffolding process; this skill provides a structured approach to classify app archetypes, fetch current OpenAI docs, and produce a minimal, working repo that is ready for local testing and submission workflows.
Core Features & Use Cases
- Classify requests into a primary app archetype (tool-only, vanilla-widget, react-widget, interactive-decoupled, submission-ready) to lock the repo shape.
- Plan tools, architectures, and upstream starting points (official OpenAI examples, ext-apps examples, or local scaffolds) tailored to the task.
- Patch or scaffold a MCP server + widget setup with tool descriptors, resource metadata, CSP, and URI versioning, aligned with repo-contract guidelines.
- Provide citations to
$openai-docs pages and a clear handoff that includes local run steps, hosting guidance, and submission-readiness if needed.
Quick Start
Run the skill with a clear app request to scaffold a minimal, working ChatGPT Apps repository and test it locally.