claude

mcp-apps-studio

Expert system for building agent-rendered interactive UI: MCP Apps (ui:// resources, the ui/* postMessage bridge), mcp-ui, the OpenAI Apps SDK (window.openai), Google A2UI, and AG-UI. 14 commands, 16 skills, 13 agents, 7 runnable project templates, and 4 scripts — scaffolder, spec validator, host-portability checker, and a local host harness — that turn any product requirement into a spec-correct, CSP-hardened, host-portable agent UI.

0 on GitHubby Markus Ahlingv1.0.0MITupdated 20 days ago
/

Plugin mcp-apps-studio from marketplace taia-marketplace in trusted-american/marketplace. Commands read from the repo's manifest — 19 days ago.

What it installs

Skills (16)

  • a2ui-protocol

    This skill should be used when building A2UI (Agent to UI) interfaces — the six streamed message types, surfaces, the flat adjacency-list component model, component catalogs, JSON Pointer data binding and two-way input binding, actions vs functionCalls, CheckRule validation, and progressive streamin

  • ag-ui-protocol

    This skill should be used when integrating AG-UI — the full event taxonomy (lifecycle, text message, tool call, state, activity, reasoning, custom), RunAgentInput, SSE and binary transports, AbstractAgent/HttpAgent, STATE_DELTA JSON Patch, generative UI, human-in-the-loop interrupts, and CopilotKit

  • host-capability-matrix

    This skill should be used when deciding which MCP Apps or window.openai APIs are safe to depend on for a given host — the verified M365 Copilot support matrix for the component bridge, tool descriptor _meta, annotations, resource _meta, CSP properties, and client _meta, plus the feature-detection pa

  • m365-copilot-packaging

    This skill should be used when shipping an MCP app into Microsoft 365 Copilot as a declarative agent — prerequisites, the widget host CORS URL, OAuth 2.1 and Entra SSO redirect URIs, Agents Toolkit sideloading, dynamic tool discovery, and the Copilot-specific capability gaps to design around.

  • mcp-apps-protocol

    This skill should be used when implementing or debugging the MCP Apps extension — ui:// resource URIs, the text/html;profile=mcp-app mimeType, every _meta.ui field, the full ui/* JSON-RPC bridge over postMessage, tool visibility, host context, display modes, and the five lifecycle phases.

  • mcp-apps-sdk

    This skill should be used when writing code against @modelcontextprotocol/ext-apps — registerAppTool, registerAppResource, RESOURCE_MIME_TYPE, the App class and its handlers, the React useApp hook, the app-bridge entry point, and the Vite single-file build that MCP Apps expect.

  • mcp-ui-sdk

    This skill should be used when working with mcp-ui — createUIResource and its rawHtml/externalUrl/remoteDom content types, the five UI action types (tool, prompt, link, intent, notify), the messageId async response flow, iframe lifecycle events, and the client-side UIResourceRenderer/AppRenderer com

  • openai-apps-sdk

    This skill should be used when building ChatGPT apps with the OpenAI Apps SDK — the window.openai bridge, openai/* _meta fields and their MCP Apps equivalents, widget state, display modes, the decoupled data/render tool pattern, checkout, file handling, and the esbuild component bundle.

  • protocol-selection

    This skill should be used when choosing between MCP Apps, mcp-ui, the OpenAI Apps SDK, A2UI, and AG-UI for a given requirement — the decision tree, the axes that actually discriminate, valid combinations, anti-patterns, and how to keep a migration path open.

  • tool-metadata-design

    This skill should be used when designing the tool surface behind an agent UI — action-oriented naming, descriptions that drive correct model selection, input and output schemas, structuredContent vs content vs _meta, safety annotations, tool visibility, and the discovery metadata that decides whethe

  • ui-performance

    This skill should be used when an agent UI is slow, heavy, or janky — bundle size and single-file inlining, skeletons from tool-input, ontoolinputpartial streaming, avoiding remount via decoupled tools, polling and chunked transfer through app-only tools, offscreen pausing with IntersectionObserver,

  • ui-porting-migration

    This skill should be used when migrating an agent UI between protocols or SDKs — OpenAI Apps SDK to portable MCP Apps, mcp-ui legacy actions to the ui/* bridge, MCP Apps to A2UI, adding AG-UI to an existing app, the adapter layer that makes ports cheap, and the field-by-field mapping tables.

  • ui-security-sandbox

    This skill should be used when hardening an agent UI — authoring the narrowest _meta.ui.csp allowlists, the iframe sandbox and proxy-origin model, treating structuredContent as untrusted input, prompt-injection through rendered data, host-mediated navigation, secrets handling, and the pre-ship secur

  • ui-state-architecture

    This skill should be used when deciding where state lives in an agent UI — the three ownership tiers (authoritative business data, ephemeral UI state, durable cross-session state), the write-through loop, updateModelContext vs setWidgetState, remount survival, and AG-UI shared-state reconciliation.

  • ui-testing-harness

    This skill should be used when testing or debugging an agent UI locally — the basic-host loop, MCP Inspector, cloudflared tunnels into Claude or ChatGPT, sendLog, the bridge-level assertions worth writing, and a triage table mapping every common symptom to its cause.

  • widget-ux-patterns

    This skill should be used when designing the visual and interaction layer of an agent-rendered widget — choosing among inline card, carousel, fullscreen, and picture-in-picture, action budgets, layout and typography rules, host theming, accessibility requirements, and the vendor dos and donts.

Slash commands (14)

  • /mcp-apps-studio:a2ui

    Build agent-driven UI as data — design a component catalog, emit createSurface/updateComponents/updateDataModel streams, wire JSON Pointer bindings and CheckRule validation, and validate against the three A2UI schemas

  • /mcp-apps-studio:agui

    Wire a long-running agent to a frontend — RunStarted through RunFinished, streamed text and tool calls, StateSnapshot/StateDelta, reasoning and activity events, interrupts, and CopilotKit or a custom subscriber

  • /mcp-apps-studio:audit

    One pass over everything — spec conformance, decoupled tools, bridge ordering, CSP, untrusted-payload handling, accessibility, bundle budget, and per-host portability — with findings ranked blocking first

  • /mcp-apps-studio:bridge

    Get the View talking to the host correctly — handler ordering, lifecycle coverage, feature-detected capabilities, and a single adapter module so components never touch a host API directly

  • /mcp-apps-studio:component

    Build the widget — picks the display mode, writes all four states, wires host theme and safe areas, enforces WCAG AA and keyboard access, and keeps every host-dependent affordance feature-detected

  • /mcp-apps-studio:copilot

    Get a widget rendering in M365 Copilot — generate the widget host CORS URL, register OAuth 2.1 and Entra SSO redirect URIs, sideload via Agents Toolkit, and design around every unsupported API

  • /mcp-apps-studio:csp

    Compute the minimal connectDomains and resourceDomains from actual code, place CSP where the host reads it, and audit for untrusted-payload rendering, unsafe URLs, injection into model context, and leaked secrets

  • /mcp-apps-studio:new

    Turn a sentence into a working project — server, View, build config, CSP, and a local test loop — for MCP Apps, mcp-ui, OpenAI Apps SDK, A2UI, or AG-UI

  • /mcp-apps-studio:port

    Move between OpenAI Apps SDK, MCP Apps, mcp-ui, A2UI, and AG-UI — introduces the adapter layer, applies the field mappings, dual-writes aliases during rollout, and flags every non-portable capability

  • /mcp-apps-studio:preview

    See the widget actually render — Inspector for protocol shape, basic-host for the real bridge, cloudflared into Claude or ChatGPT, plus a symptom-to-cause triage table

  • /mcp-apps-studio:protocol

    Decide which agent-UI protocol fits — asks the three questions that actually discriminate, then recommends a stack with reasoning, rejected alternatives, and the constraints that would change the answer

  • /mcp-apps-studio:ship

    The full loop — choose the protocol, scaffold, design the tool surface, build the View, harden CSP, audit against every target host, and verify it renders in each one before calling it done

  • /mcp-apps-studio:state

    Place every value in the right tier — authoritative server data, ephemeral View state, durable cross-session storage — wire the write-through loop, and stop losing selections on remount

  • /mcp-apps-studio:tool

    Make the model actually call your widget — action-oriented names, trigger-situation descriptions, constrained schemas, decoupled data/render tools, and the structuredContent/content/_meta split

Subagents (13)

  • mcp-apps-studio:a2ui-catalog-designer

    Use this agent to author A2UI component catalogs, surface message sequences, JSON Pointer data models, and CheckRule validation, and to validate them against common_types.json, agent_to_renderer.json, and the catalog schema. Writes catalog and stream definitions.

  • mcp-apps-studio:agui-stream-engineer

    Use this agent to wire a long-running agent to a frontend over AG-UI — emitting the correct event sequence, encoding SSE, handling StateSnapshot and StateDelta, preserving rawEvent when wrapping a framework, and adding interrupts. Writes server and client code.

  • mcp-apps-studio:bridge-engineer

    Use this agent to build or fix the View-to-host bridge — the ui/* JSON-RPC layer, window.openai extensions, mcp-ui actions, handler ordering, postMessage source validation, feature detection, and the HostBridge adapter that keeps components portable. Writes code.

  • mcp-apps-studio:csp-security-auditor

    Use this agent to run the security review on an agent UI — compute the narrowest connectDomains and resourceDomains from actual code, verify CSP placement, and find innerHTML on untrusted payloads, unvalidated URLs, raw third-party text entering model context, and secrets in the bundle. Read-only; i

  • mcp-apps-studio:host-compat-auditor

    Use this agent to verify an agent UI is portable across its target hosts — cross-references every window.openai and app.* call, _meta field, annotation, and CSP property against the per-host support matrices, and reports each gap with its design consequence rather than just a flag. Read-only.

  • mcp-apps-studio:mcp-app-builder

    Use this agent to build a working MCP App from a specification — registerAppTool and registerAppResource wiring, the View with its bridge and boundary validation, the single-file build, and a verified render in the local host harness. Writes code.

  • mcp-apps-studio:tool-schema-architect

    Use this agent to design or repair MCP tool metadata for a UI-bearing server — action-oriented names, descriptions that state trigger situations and prerequisites, constrained schemas, the decoupled data/render split, tool visibility, and the structuredContent/content/_meta division. Writes server c

  • mcp-apps-studio:ui-accessibility-reviewer

    Use this agent to audit an agent widget's accessibility — contrast ratios, alt text, keyboard reach and focus order, ARIA roles and live regions, text resizing, reduced motion, and the accessibility attributes A2UI renderers must plumb through. Read-only.

  • mcp-apps-studio:ui-perf-optimizer

    Use this agent when an agent widget is slow, heavy, or janky — bundle budget and single-file inlining, skeletons from tool input, eliminating remounts, round-trip batching, polling cleanup, offscreen pausing, and size-change debouncing. Writes code.

  • mcp-apps-studio:ui-port-migrator

    Use this agent to port an agent UI between the OpenAI Apps SDK, MCP Apps, mcp-ui, A2UI, and AG-UI — it establishes the HostBridge adapter first, dual-writes legacy aliases during rollout, and reports every non-portable capability rather than dropping it. Writes code.

  • mcp-apps-studio:ui-protocol-strategist

    Use this agent to decide between MCP Apps, mcp-ui, the OpenAI Apps SDK, A2UI, and AG-UI for a stated requirement. Returns a recommendation with reasoning, rejected alternatives, host constraints, and the conditions that would flip the decision. Read-only — it advises, it does not write code.

  • mcp-apps-studio:ui-state-architect

    Use this agent to design or repair state ownership in an agent UI — the three tiers, the write-through loop, remount survival, what reaches model context and when, and AG-UI snapshot/delta reconciliation. Writes code.

  • mcp-apps-studio:widget-designer

    Use this agent to design or improve what the user actually sees in an agent widget — display mode selection, action budgets, all four content states, host theming and safe areas, and the vendor layout and typography rules. Writes View code.

View source on GitHub

More from trusted-american/marketplace

All 39 plugins →

Unified delivery orchestration: one plugin replacing six. A tested, dependency-free kernel (swarm conductor with leased claims and worktree isolation, guardrails, transactional state MCP, fleet protocol) behind platform adapters for Linear, Jira, GitHub and Harness. 133 commands, 156 agents, 52 skills, 12 schema-validated workflows, 155 tests.

52 skills · 133 commands · 156 agents · hooks

0

Modern Claude Code second brain: 5-layer stack deploy + three-tier memory (engram + Obsidian vault + plugin rules) + 22-tool MCP reference server. 21 behavior-triggering skills, 12 single-intent commands, 18 role-scoped agents bridging engram and Obsidian. Tracks current Claude Code capabilities (Fable 5 + Opus 4.8, agent teams, Tool Search, web/remote).

70 skills · 12 commands · 18 agents

0

a3-plugin

claude

Fullstack development agent for the A3 platform — orchestrates Ember.js, Firebase/Firestore, and GCP Cloud Functions with deep codebase knowledge, context-budgeted delegation, targeted review, and CI-based verification

50 skills · 11 commands · 11 agents · 2 MCP servers

0

Comprehensive Material UI (MUI) expert plugin — 26 skills, 14 commands, 7 agents covering theming, CSS variables, Pigment CSS, components, sx/styled, slots API, MUI X (DataGrid, DatePickers, Charts, TreeView), accessibility, performance, SSR/Next.js, animations, virtualization, forms, white-label/multi-tenant, headless (MUI Base), Joy UI, i18n/RTL, testing, migration, entity-driven CRUD, ecosystem integrations, and 200+ creative widget patterns.

27 skills · 14 commands · 7 agents

0

Consul - Enterprise Microsoft ecosystem orchestrator for TVS (Trusted Virtual Solutions) multi-entity multi-tenant deployment. 19 agents, 18 commands, 17 skills, 14 hooks, 5 workflows. MCP server for Graph API, Dataverse, Fabric, and Planner.

1 skill · 18 commands · 19 agents

0

Comprehensive .NET 10, Blazor, ASP.NET Core, C#, and Syncfusion expert plugin for building modern web apps, microservices, and cloud-native solutions

19 skills · 8 commands · 6 agents · hooks

0

Similar plugins

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

mempalace

claudecodex

AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.

3 skills · 6 commands · 1 MCP server · hooks

MemPalace/mempalace 60,536

Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer

8 skills · 1 MCP server

ChromeDevTools/chrome-devtools-mcp 55,181

gitnexus

claudecodex

Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.

12 skills · 1 MCP server · hooks

abhigyanpatwari/GitNexus 48,482

daisyUI Blueprint MCP server - (Needs paid license)

saadeghi/daisyui 42,821

Maintain mcp-apps-studio?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins39 plugins