> ## Documentation Index
> Fetch the complete documentation index at: https://docs.backstack.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Platform Overview

> Understand Backstack's core concepts and capabilities

# Platform Overview

Backstack is an AI toolkit that helps organizations deploy, manage, and secure AI tools and resources across their teams.

## What is Backstack?

Backstack provides a centralized platform for managing AI infrastructure at the organizational level. Instead of individuals managing their own AI tools and API keys, organizations can:

* **Install tools once**, make them available to everyone
* **Control access** with role-based permissions and security policies
* **Monitor usage** across teams with analytics and audit logs
* **Secure AI interactions** with compliance policies and client restrictions

Backstack integrates with any AI client that supports the Model Context Protocol (MCP), including Claude Desktop, VS Code, Cursor, Windsurf, and Zed.

## Core Concepts

### Organizations

An organization is your team's workspace in Backstack. It provides:

* Team member management with role-based access control
* Shared AI tools and services available to all members
* Security policies and compliance controls
* Usage analytics and audit trails

Learn more: [Organization Management](/organizations)

### Workspaces

Workspaces organize tools, resources, and team members into logical groups:

* **Tools** - Executable actions and integrations (MCP servers)
* **Resources** - Data sources and context (documents, databases)
* **Members** - Team members with specific access permissions
* **Settings** - Visibility, permissions, and configuration

Workspaces help teams collaborate on specific projects or workflows while maintaining security boundaries.

Learn more: [Workspaces](/workspaces)

### MCP (Model Context Protocol)

MCP is an open protocol that connects AI clients to tools and resources. Backstack acts as an MCP server, providing your AI clients with access to:

* Organization-installed tools (npm packages and remote servers)
* Shared documents and knowledge bases
* Security-validated operations

When you connect an AI client to Backstack via MCP, all your organization's tools become available in your conversations.

Learn more: [MCP Setup](/mcp)

### Tools & Services

Tools are the actions your AI can perform. In Backstack, tools come from:

* **NPM packages** - MCP servers installed from npm and run automatically
* **Remote servers** - External MCP servers connected via URL

Organization admins install tools once, and they become available across workspaces.

Learn more: [Organization Services](/organization-services)

### Resources

Resources provide context to your AI:

* **Documents** - PDFs, Word docs, text files shared to workspaces
* **Data sources** - Databases, APIs, and other information sources

Resources are discovered from your desktop and shared to specific workspaces for AI access.

Learn more: [Documents](/documents)

## Key Features

<CardGroup cols={2}>
  <Card title="AI Chat" icon="message-bot" href="/chat">
    Have conversations with AI assistants that can execute tools and access resources
  </Card>

  <Card title="Security Policies" icon="shield-check" href="/security-policies">
    Define natural language security rules that validate tool execution in real-time
  </Card>

  <Card title="Activity Logs" icon="list" href="/activity-logs">
    Complete audit trail of all actions across your organization
  </Card>

  <Card title="Analytics" icon="chart-bar" href="/metrics">
    Track tool usage, performance, and adoption across your team
  </Card>

  <Card title="Desktop App" icon="desktop" href="/desktop">
    Local file discovery and document sync from your desktop
  </Card>

  <Card title="Client Restrictions" icon="lock" href="/organizations#security--client-restrictions">
    Control which AI clients can access your organization
  </Card>
</CardGroup>

## Use Cases

### Development Teams

* Install GitHub, database, and filesystem MCP servers
* Share team codebases and documentation as resources
* Set security policies to prevent destructive operations
* Monitor AI tool usage across the team

### Enterprise Organizations

* Centralize AI infrastructure management
* Enforce compliance with security policies
* Audit all AI interactions for regulatory requirements
* Control which AI clients employees can use

### Research Teams

* Share research papers and datasets as searchable resources
* Install domain-specific MCP tools for data analysis
* Collaborate across workspaces for different projects
* Track AI usage and effectiveness

## How It Works

1. **Organization admins** install MCP tools and configure settings
2. **Team members** are invited and assigned to workspaces
3. **Users connect** their AI clients (Claude Desktop, VS Code, etc.) to Backstack via MCP
4. **AI conversations** have access to organization tools and resources
5. **Security policies** validate operations in real-time
6. **Activity logs** track all actions for audit and compliance

## Benefits

* **Reduce complexity** - Install tools once, use them everywhere
* **Improve security** - Centralized control over AI tool access
* **Enable collaboration** - Share resources and tools across teams
* **Maintain compliance** - Complete audit trail and policy enforcement
* **Scale AI adoption** - Make powerful AI capabilities accessible to all team members

## Getting Started

<CardGroup cols={2}>
  <Card title="Quickstart Guide" icon="rocket" href="/quickstart">
    Get up and running with Backstack in under 10 minutes
  </Card>

  <Card title="Create Your First Workspace" icon="folder" href="/workspaces">
    Learn how to organize tools and resources into workspaces
  </Card>

  <Card title="Install MCP Tools" icon="server" href="/organization-services">
    Add tools and services to your organization
  </Card>

  <Card title="Connect an AI Client" icon="plug" href="/mcp">
    Set up Claude Desktop, VS Code, Cursor, or other MCP clients
  </Card>
</CardGroup>