AI-First Console

AI is the Console

Chat is the new command line. Deploy 200+ managed open source services through natural language — no dashboards, no YAML, no clicking. Just describe what you need and your AI agent provisions real infrastructure.

Works with:
Try in Browser

Try Instantly — No Install Needed

Experience OSC + AI in your browser. No local setup, no configuration.

Our web-based AI agent gives you instant access to all 200+ OSC services. Perfect for exploring capabilities before setting up IDE integration.

Try Web Agent Now

Want local dev experience?

Choose an IDE integration below for seamless development workflow.

What You Can Build

Tiered example prompts for every skill level

Beginner

Create a PostgreSQL database called myapp
Set up a Valkey cache instance
Start a Redis server for my project
Create a MySQL database

Intermediate

Start WordPress with a MariaDB backend
Start my Node.js app from github.com/user/repo
Deploy a Next.js application with PostgreSQL
Deploy a WASM function for edge computing
Set up a monitoring stack with Prometheus and Grafana

Advanced

Build a complete VOD pipeline with storage, transcoding, and packaging
Ask the OSC Architect how to build a live streaming platform
Create a microservices architecture with message queues and service mesh
Deploy a distributed ML inference pipeline
Prompt Recipes

Copy, Paste, Deploy — Recipes for Your AI Agent

Ready-to-use prompts that build complete applications. Just paste into your AI agent and watch it work.

beginner

Build a Web Shop

Create a complete e-commerce application with product catalog, shopping cart, and PostgreSQL database.

View prompt
intermediate

Deploy a Streaming Service

Set up a complete video streaming pipeline with transcoding, packaging, and a web player.

View prompt
intermediate

Launch an AI Assistant Backend

Build a conversational AI backend with REST API, database persistence, and session caching.

View prompt
View all 9 prompt recipes
Featured Tutorial
15 min
Beginner
→ Deployed app

Build Your First App

Go from zero to a deployed full-stack application using Claude Code and Open Source Cloud.

Start the Tutorial
Vibe Deploy — From Prompt to Production

Your AI agent does not just write code. With OSC MCP, it deploys infrastructure too — databases, applications, pipelines — all through conversation. And every service is portable open source.

Describe → Deploy

Tell your AI agent what you need. It provisions real infrastructure in seconds — no YAML, no consoles, no manual steps.

Zero Lock-In

Every service is an unmodified open source project. Move to any cloud or on-premises at any time.

163+ Services

Databases, caches, CDNs, applications, monitoring, AI tools — all available through a single MCP connection.

Read the Full Story

Production-Ready, Not Sandboxed

Other tools let you prototype in a sandbox. OSC MCP deploys real infrastructure from the start.

Sandbox-then-Export
OSC MCP
Workflow
Plan → Sandbox → Export IaC → Configure CI/CD → Deploy
Describe → Deployed
Time to production
Hours / days
Minutes
Lock-in
Requires specific IaC tooling
Zero lock-in (open source)
Transparency
Black-box sandbox
Full source code access

See the proof

Streaming Tech TV+ — 14 services in 36 hours Specter CRM — full CRM in a weekend
Quick Start

Start Building with AI on OSC

One command connects your AI agent to 200+ managed open source services. Your agent provisions infrastructure as you code.

Need a token? Sign up free at app.osaas.io

Instant Infrastructure

Databases, storage, queues — deployed by describing what you need.

Deploy from GitHub

Point your agent at any repo. OSC builds, deploys, and gives you a URL.

Built-in AI Architect

An expert assistant that knows every OSC service and designs your architecture.

Production from Day One

No sandbox-to-production migration. What you build is already live.

Create Free Account

Common Questions About Building on OSC

Run a single command in your terminal: claude mcp add --transport http osc https://mcp.osaas.io/mcp. For Cursor, VS Code, or other tools, add the MCP server URL https://mcp.osaas.io/mcp in your MCP configuration with a Bearer token from your free OSC account at app.osaas.io. Once connected, your AI agent can discover and deploy any of 200+ managed open source services through natural language.

40+ MCP Tools

Access the complete OSC infrastructure through natural language

Infrastructure

Databases, storage, caching, message queues

Applications

Deploy and manage complete applications

Intelligence

AI-powered architecture planning and optimization

Ask OSC Architect

Get expert guidance on building complex architectures

Pricing for Builders

Flexible plans that grow with your projects

Free

€0

100 tokens total

Try databases for ~1 week
Access to all 200+ services
MCP integration included
Community support
Popular

Personal

€15/mo

20 tokens/day

3-5 running services
All infrastructure tools
Application hosting
Priority support

Professional

€199/mo

50 tokens/day

Production workloads
Custom domains
Advanced monitoring
Dedicated support
View Full Pricing Details|Explore Use Cases

How to Connect

Choose your preferred AI tool and follow the setup instructions below.

Claude Desktop

Connect OSC to Claude Desktop app for a seamless AI assistant experience.

Download Claude Desktop

Setup Steps:

  1. 1Download Claude for Desktop for macOS or Windows
  2. 2Go to Settings → Connectors → "Add custom connector"
  3. 3Name the connector (e.g., "OSC") and enter the MCP URL
  4. 4Click "Connect" and authenticate through the OSC web interface

Claude Code

Connect OSC with Claude Code CLI for terminal-based AI assistance.

Setup Steps:

  1. 1Prerequisite: Install Node.js 18+ from nodejs.org
  2. 2Run the add command to register the MCP server
  3. 3Start Claude Code
  4. 4Connect to the MCP using the /mcp command

Commands

Add the MCP server:

$ claude mcp add --transport http osc https://mcp.osaas.io/mcp

Start Claude Code:

$ claude

Connect to the MCP:

> /mcp
Alternative: Personal Access Token

If browser authentication fails, use a personal access token from Settings/API:

{
  "mcpServers": {
    "osc": {
      "type": "http",
      "url": "https://mcp.osaas.io/mcp",
      "headers": {
        "Authorization": "Bearer <personal-access-token>"
      }
    }
  }
}

VS Code

Integrate OSC tools directly into your VS Code development environment.

Setup Steps:

  1. 1Prerequisite: Install Node.js 18+ from nodejs.org
  2. 2Open View → Command Palette
  3. 3Search for "MCP" and select "Add server"
  4. 4Choose HTTP transport
  5. 5Enter the MCP URL and name your connector
  6. 6Authenticate when prompted

Codex

Connect Codex to leverage OSC tools in your development workflow.

Setup Steps:

  1. 1Get your Personal Access Token from app.osaas.io → Settings → API
  2. 2Export it as an environment variable: export OSC_ACCESS_TOKEN=<your-token>
  3. 3Edit ~/.codex/config.toml and add the configuration shown below
  4. 4Run codex — OSC MCP tools will be available automatically

Configuration (~/.codex/config.toml)

# ~/.codex/config.toml
[mcp_servers.osc]
url = "https://mcp.osaas.io/mcp"
bearer_token_env_var = "OSC_ACCESS_TOKEN"

The bearer_token_env_var field references the name of an environment variable — never embed your token directly in config files.

ChatGPT

Connect ChatGPT to OSC using Developer Mode for AI-powered automation.

Setup Steps:

  1. 1Go to Settings → Connectors → Advanced → Developer Mode
  2. 2Click "Add Connector"
  3. 3Enter the MCP URL and name your connector (e.g., "OSC")
  4. 4Authenticate when prompted through the OSC web interface

GitHub Actions

Give your Claude Code agent access to OSC tools in CI/CD workflows.

Setup Steps:

  1. 1Add your OSC personal access token as a GitHub Actions secret (e.g. OSC_ACCESS_TOKEN)
  2. 2Create a workflow step that writes an MCP config JSON file
  3. 3Pass the config to claude-code-action using --mcp-config
  4. 4Whitelist the OSC MCP tools in --allowedTools

Workflow Configuration (YAML)

name: Claude Code with OSC
on:
  issues:
    types: [opened, labeled]

permissions:
  contents: write
  issues: write
  pull-requests: write
  id-token: write

jobs:
  analyze:
    if: contains(github.event.issue.labels.*.name, 'claude')
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Create OSC MCP Config
        run: |
          cat > /tmp/osc-mcp-config.json << EOF
          {
            "mcpServers": {
              "osc": {
                "type": "http",
                "url": "https://mcp.osaas.io/mcp",
                "headers": {
                  "Authorization": "Bearer ${{ secrets.OSC_ACCESS_TOKEN }}"
                }
              }
            }
          }
          EOF

      - uses: anthropics/claude-code-action@v1
        with:
          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
          claude_args: |
            --mcp-config /tmp/osc-mcp-config.json
            --allowedTools "mcp__osc__list-available-services,\
              mcp__osc__get-service-schema,\
              mcp__osc__get-service-endpoints,\
              mcp__osc__list-my-services,\
              mcp__osc__list-service-instances,\
              mcp__osc__describe-service-instance,\
              mcp__osc__create-service-instance,\
              mcp__osc__delete-service-instance,\
              mcp__osc__call-service-endpoint,\
              mcp__osc__get-logs-for-instance,\
              mcp__osc__create-database,\
              mcp__osc__create-storage-bucket,\
              mcp__osc__list-objects-on-bucket,\
              mcp__osc__ask-osc-architect,\
              mcp__osc__get-mcp-help"

Store your OSC personal access token as a repository secret. Never commit tokens directly in workflow files.

Cursor

Use OSC services directly in Cursor IDE with AI-powered coding assistance.

Setup Steps:

  1. 1Get your Personal Access Token from app.osaas.io → Settings → API
  2. 2Create or edit .cursor/mcp.json (project-level) or ~/.cursor/mcp.json (global) and add the configuration shown below
  3. 3Restart Cursor — OSC MCP tools will appear automatically in agent mode

Configuration (.cursor/mcp.json)

// .cursor/mcp.json (project) or ~/.cursor/mcp.json (global)
{
  "mcpServers": {
    "osc": {
      "url": "https://mcp.osaas.io/mcp",
      "headers": {
        "Authorization": "Bearer <your-personal-access-token>"
      }
    }
  }
}

Get your personal access token from app.osaas.io → Settings → API. Replace <your-personal-access-token> with your actual token.

Windsurf

Integrate OSC into Windsurf IDE for seamless AI-assisted development.

Setup Steps:

  1. 1Get your Personal Access Token from app.osaas.io → Settings → API
  2. 2Edit ~/.codeium/windsurf/mcp_config.json and add the configuration shown below
  3. 3Restart Windsurf — OSC MCP tools will appear automatically in agent mode

Configuration (~/.codeium/windsurf/mcp_config.json)

// ~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "osc": {
      "serverUrl": "https://mcp.osaas.io/mcp",
      "headers": {
        "Authorization": "Bearer <your-personal-access-token>"
      }
    }
  }
}

Get your personal access token from app.osaas.io → Settings → API. Replace <your-personal-access-token> with your actual token.

Host Your MCP Server on OSC

Your MCP server, production-ready in minutes. HTTPS, SSE, auto-scaling — no DevOps required.

Deploy from GitHub

Any Node.js, Python, or Wasm MCP server. Just point us to your repo.

HTTPS Endpoint Instantly

Get a public URL like https://{app}.auto.prod.osaas.io/mcp — share it with anyone.

Auto-Scaling, Zero Lock-In

Your code stays yours. Deploy anywhere, no vendor lock-in.

Deploy Your MCP Server |Read the Tutorial

Connect Your AI Agent in 2 Minutes

Connect to Open Source Cloud today and unlock the full potential of open source media tools with your favorite AI assistant.

Try in BrowserRead Full Documentation
Try: "Create a PostgreSQL database" with your AI assistant

Last updated: March 2026