Services & SDKs

GenMB includes 12 core built-in services (documented below) plus 40+ specialized SDKs that add backend capabilities to your generated apps without any configuration. The AI analyzes your code, detects which services are needed, and automatically injects the corresponding client SDKs.

Services are detected automatically, no manual setup or API keys required. Each service has a client SDK injected as a script tag into your app's HTML. Services work both in preview (via Next.js proxy) and in deployed apps (via deploy-service proxy). All services are available on the Free plan unless noted otherwise.

Cloud file storage for your generated apps. Upload, list, and delete files via window.genmb.storage. Backed by Google Cloud Storage with plan-based quotas.

No API keys needed, the SDK authenticates through GenMB's proxy automatically.

Google OAuth authentication for your generated apps. Users can sign in with their Google account via window.genmb.auth.

The service provides a complete Google OAuth flow - sign in, sign out, and session management. User profile data (name, email, avatar) is available after authentication, and session state persists across page reloads.

AI completion proxy for building chatbot interfaces and text generation features. Available via window.genmb.ai. Routes through GenMB's backend so you never need to handle API keys.

Send prompts and receive AI-generated completions through a simple API. The service supports streaming responses for real-time chat interfaces. AI proxy credits are charged to the app owner, not end-users of the deployed app.

Text-focused AI completions for summarization, translation, transformation, and structured generation. Detected as a separate capability from chatbot, the AI auto-injects when your app needs one-shot text operations rather than turn-by-turn chat.

Uses the same proxy as the AI Chatbot SDK and shares the 100 requests/hour rate limit. Each completion deducts 1 credit from the app owner's balance.

Generate images from text prompts inside your generated apps. Detected when your app references image generation, AI artwork, or thumbnail creation.

Routes through GenMB's AI proxy. 1 credit per image generation, charged to the app owner. Available on the Free plan with the standard 100 req/hour AI rate limit.

Generate short videos from text prompts using Sora 2. API-only, no client SDK; integrate via the proxy endpoint your app calls server-side.

Requires the Pro or Business plan. Each generation costs 5 credits and is charged to the app owner. Rate limited to 5 generations per hour and 20 per day per app. See the dedicated Video Generation docs for prompt patterns and status polling.

Realtime voice chat for your generated apps via window.genmb.voice. Backed by Azure OpenAI Realtime (gpt-realtime-1.5) - bidirectional audio over WebRTC with sub-second latency.

Add a voice button or always-on voice surface to your app and the SDK handles microphone capture, streaming, and playback. Each session costs 5 credits and is charged to the app owner. Rate limited to 20 sessions per hour per app. See the dedicated Voice Agent docs for setup.

A form submission handler for contact forms, feedback forms, and any user input that needs to be stored. Available via window.genmb.contact.

Submit form data with a single function call, no backend code needed. Submissions are stored in Firestore and accessible via the Database Browser. The service supports custom fields, so any form structure is accepted. Optional email notification is available when a new submission is received.

Role-Based Access Control for managing user permissions in your generated apps. Available via window.genmb.rbac.

Define roles (admin, editor, viewer, etc.) and assign them to users. Check permissions before showing UI elements or executing actions. Roles are stored per-app and managed through the SDK.

Client SDK for reading and writing data to your app's database. Available via window.genmb.data. Supports both Firestore and PostgreSQL via DataConnect.

Query, create, update, and delete records in your app's database. Works with Firestore collections for NoSQL data access and with PostgreSQL tables when using DataConnect for relational data.

Semantic search and RAG (Retrieval-Augmented Generation) for your generated apps. Available via window.genmb.vectordb. Backed by pgvector with automatic text chunking and embedding.

Ingest documents, web pages, or any text content and search them semantically using natural language queries. The SDK handles chunking, embedding generation (Gemini native, 768 dimensions), and similarity search automatically. Ideal for building knowledge bases, FAQ systems, and AI-powered search features.

Embeddable AI assistant chat widget for your generated apps. Available via window.genmb.assistant. Connect to a GenMB Assistant to provide an interactive chatbot experience.

Add a floating chat widget that connects to your configured GenMB Assistant. Supports conversation history, knowledge base lookups, and scheduled tasks. The widget can be customized with your branding and positioned anywhere in the app.

GenMB's AI automatically analyzes your generated code to detect which services are needed. Matching SDKs are injected into your app's HTML without manual configuration.

After code generation, the AI scans for patterns like file upload forms, login buttons, chat interfaces, and data operations. Each service has detection hints defined in its capability definition (e.g., "file upload", "sign in with Google", "send message").

Detected services are listed in the Services panel in the app editor sidebar. You can manually enable or disable services from the Services panel if the auto-detection is not what you want. SDKs are re-injected on every code save to stay in sync with your app's current functionality.

Services have plan-based quotas to ensure fair usage. Here are the key limits for the most commonly used services.

FAQs