butterbase

Set up a Postgres-backed backend with auth, storage, serverless functions, and MCP AI integration.

3.0k|148|Updated May 20, 2026
One-click install
npx skills add https://github.com/butterbase-ai/butterbase --skill butterbase
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: butterbase
Source: https://github.com/butterbase-ai/butterbase/tree/main
Command: npx skills add https://github.com/butterbase-ai/butterbase --skill butterbase

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node.js, docker, postgresql, redis, localstack, deno, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a comprehensive solution for building AI-driven applications without lock-in, offering a Postgres-backed backend with various services like row-level security, serverless functions, an AI gateway, and more.

Core Features & Use Cases

  • Postgres Data Plane: Offers per-app databases with declarative schemas, automatic REST endpoints, and migrations.
  • Row-Level Security: Provides first-class RLS policy management with user-isolation helpers.
  • Key-Value Store: Includes regional, quota-protected KV with TTL, audit trail, and dashboard expose rules.
  • File Storage: Offers S3/R2-backed object storage with presigned URLs, ACLs, and async indexing.
  • Serverless Functions: Executes TypeScript functions on the Deno runtime.
  • Durable Objects: Allows stateful per-key actors for chat rooms, multiplayer, rate limiters, and long-running agents.
  • Realtime Subscriptions: Enables WebSocket subscriptions to table changes for live UIs and presence.
  • Edge SSR: Allows deploying Next.js/Remix/Astro edge handlers from source.
  • Frontend Hosting: Offers zip or build-from-source static/SPA deploys with custom domains.
  • AI Gateway: Provides a single endpoint for chat, embeddings, model listing; pluggable router adapters.
  • RAG: Manages collections, document ingestion, semantic search, and synthesized answers.
  • Integrations: Allows third-party tool access via Composio.
  • Auth: Offers email + OAuth (Google, GitHub, Apple, X, …), JWT tuning, post-login hooks, service keys.
  • Audit Logs: Provides structured request audit trail across KV and other surfaces.
  • Webhooks: Allows outbound webhooks for app events.
  • Multi-region App Moves: Enables relocating an app across regions with retained source replicas.
  • MCP Server: Exposes every capability as MCP tools at /mcp (HTTP) or via stdio.
  • Claude Code Plugin: Provides 30+ guided skills for AI-driven app building.
  • Use Case: Imagine you are building an AI-driven chatbot app. Use this Skill to set up the backend with databases, authentication, and AI capabilities, all with the power of the Model Context Protocol (MCP).

Quick Start

Initialize an app using the butterbase skill and specify the app name.

Frequently Asked Questions about butterbase

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build an AI-driven backend with Postgres and the Model Context Protocol?

To build an AI-driven backend with Postgres and the Model Context Protocol, you can use a backend-as-a-service solution that exposes database management, authentication, and AI gateway features as MCP tools over HTTP or stdio.

Can I run serverless TypeScript functions and manage RLS policies in a single backend service?

Yes, a backend-as-a-service solution can execute TypeScript functions on the Deno runtime while providing first-class Postgres row-level security policy management with user-isolation helpers.

What do I need to set up a local backend environment with Redis, LocalStack, and Docker?

Setting up a local backend environment with Redis, LocalStack, and Docker requires installing Node.js and configuring the necessary backend services to orchestrate containers, databases, and object storage.

How does the Model Context Protocol server expose database and storage capabilities to AI agents?

The Model Context Protocol server exposes database, key-value store, file storage, and AI gateway capabilities as standardized MCP tools at the `/mcp` HTTP endpoint or via stdio for AI agent consumption.

What is the best way to integrate RAG and semantic search into an existing Postgres database?

The best way to integrate RAG and semantic search into a Postgres database is using a backend service that manages document ingestion, semantic search, and synthesized answers through a pluggable AI gateway router.

Does this backend solution support multi-region app relocation with retained source replicas?

Yes, this backend solution supports multi-region app moves, enabling you to relocate an application across regions while retaining source replicas for data continuity.