Skybridge
Ship a ChatGPT or MCP app with SPEC-driven architecture, tools, shared views, and deployment to ChatGPT.
Overview
Skybridge is an agent skill most often used in Build (also Validate scope, Ship launch) that guides solo builders through SPEC-first ChatGPT and MCP app creation with Skybridge.
Install
npx skills add https://github.com/alpic-ai/skybridge --skill skybridgeWhat is this skill?
- Full lifecycle: discover, SPEC.md, architecture, bootstrap, implement, debug, local dev, deploy, connect to ChatGPT
- Dual-user model: human interacts with the view while the assistant LLM reads shared state
- Hard gate: no implementation without SPEC.md; architecture pass updates SPEC before coding
- Reference-driven flow: copy-template, run-locally, implement-tools, implement-views, debugging, deploy
- Skybridge framework guidance for MCP apps consumed inside ChatGPT conversations
Adoption & trust: 514 installs on skills.sh; 1.7k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You want a ChatGPT or MCP app but lack a repeatable path from idea to SPEC, shared views, tools, and deployed connection.
Who is it for?
Solo builders shipping MCP servers or ChatGPT apps who need UX guidelines and a gated SPEC → implement → deploy checklist.
Skip if: Teams that already have frozen specs and only need a one-off REST API with no conversational UI or MCP surface.
When should I use this skill?
User wants to create or update a ChatGPT app, MCP app, MCP server, or use the Skybridge framework.
What do I get? / Deliverables
You maintain an up-to-date SPEC.md, implement MCP tools and views against architecture references, and deploy a app the assistant can invoke in ChatGPT.
- Updated SPEC.md
- MCP server with tools and views
- Deployed app connected to ChatGPT
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Building conversational MCP servers and Skybridge UI is core product construction for LLM-facing apps. Agent-tooling is where skills teach MCP servers, tools, and assistant-consumed views—not generic CRUD backends.
Where it fits
Brainstorm ChatGPT app ideas against UX guidelines and capture decisions in SPEC.md before bootstrapping.
Implement MCP tools that the assistant invokes during a conversation after architecture is documented.
Build shared views the human clicks while the LLM reads the same component state.
Run the dev server, debug failures, deploy, and connect the app to ChatGPT for end-to-end testing.
How it compares
Use instead of piecing together generic MCP server tutorials without ChatGPT view UX or lifecycle gates.
Common Questions / FAQ
Who is skybridge for?
Indie developers and small teams building ChatGPT apps, MCP servers, or Skybridge-based conversational experiences with tools and custom views.
When should I use skybridge?
When starting or updating a ChatGPT/MCP app in Validate (brainstorm against UX guidelines), Build (bootstrap and implement), or Ship (debug, dev server, deploy and connect to ChatGPT).
Is skybridge safe to install?
Review the Security Audits panel on this Prism page and inspect reference scripts and deploy steps in the skill repo before granting network or deploy permissions.
SKILL.md
READMESKILL.md - Skybridge
# Creating Apps For LLMs Those are conversational experiences that extend AI assistants through tools and custom UI views. They're built as MCP servers invoked during conversations. ⚠️ The app is consumed by two users at once: the **human** and the **AI Assistant LLM**. They collaborate through the view—the human interacts with it, the LLM sees its state. Internalize this before writing code: the view is your shared surface. SPEC.md keeps track of the app's requirements and design decisions. Keep it up to date as you work on the app. **No SPEC.md?** → Read [discover.md](references/discover.md) first. Nothing else until SPEC.md exists. **SPEC.md exists?** → Read SPEC.md, then follow [architecture.md](references/architecture.md) to design the change. Update SPEC.md, then read the relevant Implementation references below before writing code. ## Setup 1. **Copy template** → [copy-template.md](references/copy-template.md): when starting a new project with ready SPEC.md 2. **Run locally** → [run-locally.md](references/run-locally.md): when ready to test, need dev server or ChatGPT/Claude connection ## Architecture Design or evolve UX flows and API shape → [architecture.md](references/architecture.md) ## Implementation - **Fetch and render data** → [fetch-and-render-data.md](references/fetch-and-render-data.md): when implementing server handlers and view data fetching - **State and context** → [state-and-context.md](references/state-and-context.md): when persisting view UI state and updating LLM context - **Prompt LLM** → [prompt-llm.md](references/prompt-llm.md): when view needs to trigger LLM response - **UI guidelines** → [ui-guidelines.md](references/ui-guidelines.md): display modes, layout constraints, theme, device, and locale - **External links** → [open-external-links.md](references/open-external-links.md): when redirecting to external URLs or setting "open in app" target - **Download file** → [download-file.md](references/download-file.md): when saving content to the user's filesystem - **OAuth** → [oauth.md](references/oauth.md): when tools need user authentication to access user-specific data - **CSP** → [csp.md](references/csp.md): when declaring allowed domains for fetch, assets, redirects, or iframes ## Deploy - **Ship to production** → [deploy.md](references/deploy.md): when ready to deploy via Alpic - **Publish to ChatGPT/Claude Directories** → [publish.md](references/publish.md): when ready to submit for review Full API docs: [https://docs.skybridge.tech/api-reference.md](https://docs.skybridge.tech/api-reference.md)