Pricing

Free for individuals. Predictable pricing for teams of every size.

Free

$0
  • 3 members
  • 5 integrations
  • 1,000 API calls/mo
  • 2 policies
  • 7 days audit retention
Get started

Starter

$12 /mo
  • 5 members
  • 10 integrations
  • 5,000 API calls/mo
  • 5 policies
  • 14 days audit retention
Start trial
Most Popular

Team

$29 /mo
  • 10 members
  • Unlimited integrations
  • 25,000 API calls/mo
  • Unlimited policies
  • 30 days audit retention
Start trial

Business

$99 /mo
  • 50 members
  • Unlimited integrations
  • 100,000 API calls/mo
  • Unlimited policies
  • ✓ SSO
  • 90 days audit retention
  • ✓ Dedicated runtime
Start trial

Enterprise

Custom
  • Unlimited members
  • Unlimited integrations
  • Unlimited API calls/mo
  • Unlimited policies
  • ✓ SSO
  • 365 days audit retention
  • ✓ Dedicated runtime
Contact sales

Feature comparison

FreeStarterTeamBusinessEnterprise
Members351050Unlimited
Integrations510UnlimitedUnlimitedUnlimited
API calls / month1,0005,00025,000100,000Unlimited
Policies25UnlimitedUnlimitedUnlimited
SSO
Audit retention7 days14 days30 days90 days365 days
Dedicated runtime
Context optimization
Encrypted credential vault

Frequently asked questions

What is the MCP protocol?

MCP (Model Context Protocol) is an open standard for connecting AI assistants to external tools and data sources. It lets AI clients like Cursor and Claude Desktop access your GitHub repos, databases, monitoring tools, and more through a unified interface.

How is Switchboard different from the open-source version?

The open-source Switchboard runs as a local binary for individual developers. Switchboard Hosted adds multi-tenancy, team management, encrypted credential vaults, role-based access policies, audit logging, and a web dashboard — everything you need to deploy MCP for a team or organization.

What AI clients are supported?

Any MCP-compatible client works: Cursor, Claude Desktop, Windsurf, and others. Switchboard exposes a standard MCP endpoint that any compliant client can connect to.

How are credentials stored?

All integration credentials are encrypted with AES-256-GCM before storage. The encryption key never leaves your server. OAuth tokens are stored the same way, and you can rotate them at any time.

What does context optimization mean?

LLMs have limited context windows. Switchboard automatically compacts API responses before they reach the AI — stripping redundant fields and normalizing data. This typically reduces payload size by 90%, letting you fit more useful context in every request.

Can I upgrade or downgrade my plan?

Yes. You can change plans at any time from the dashboard. Upgrades take effect immediately. Downgrades apply at the end of your billing cycle.

What happens if I exceed my API call limit?

You'll receive a warning at 80% usage. Once you hit the limit, MCP requests return an error until the next billing cycle or until you upgrade.