How to Use the Cerbos MCP in Claude Code

Q: How do I install the Cerbos MCP server in Claude Code?

Run claude mcp add --transport http cerbos-mcp -- in your terminal. Make sure all flags come before the server name. Verify the installation by running claude mcp list.

Q: Can Claude Code evaluate single AuthZEN actions with Cerbos?

Yes, it uses the authzenevaluation tool for single-action checks. You pass the principal and resource details, and the CLI returns the evaluation based on the AuthZEN entity model.

Q: How do Cerbos and Claude Code handle batch permissions?

Your terminal agent calls authzenevaluations to run multiple checks at once. It supports standard batch semantics like permitonfirstpermit, making it perfect for automated testing pipelines.

Q: Can I fetch policy configuration data from the command line?

The agent executes getauthzenconfig to return your endpoint URLs. It also uses getserverinfo to pull version and build details directly into your terminal session.

Connect Claude Code to the Cerbos MCP Server and manage authorization policies directly from your terminal. Evaluate rules headless.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Cerbos MCP to Claude Code

Create your Vinkius account to connect Cerbos to Claude Code and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Cerbos with Claude Code

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Headless Cerbos MCP Server evaluations

Terminal users connect to your authorization layer and run the `check_resources` tool to evaluate permissions instantly. You type the command, and it spits out the read-only evaluation right in your shell. SREs and DevOps engineers can pipe these results into other scripts. If you are debugging a production access issue, your terminal agent fetches the exact permission state for any principal in seconds.

Batch process AuthZEN requests via CLI

The `authzen_evaluations` tool processes multiple access requests using execute_all or deny_on_first_deny semantics directly from the command line. Running CI/CD checks on your policy changes requires reliable batch testing. You can script Claude Code to run these evaluations as part of a pre-commit hook or a GitHub Action. It catches broken access rules before they ever reach your staging environment.

Extract query plans and server config

Grabbing the right filter logic for a database query happens through the `plan_resources` tool. The agent produces the exact query plan required for a principal to access specific resources. Need to verify your deployment? The `get_server_info` and `get_authzen_config` tools return your instance build details and AuthZEN API endpoints. Everything happens through standard input and output streams.

Setup guide

Set up Cerbos MCP in Claude Code

Prerequisites

Claude Code CLI installed (npm install -g @anthropic-ai/claude-code)
Active Vinkius subscription with a valid endpoint token

1

Run the add command

Open your terminal and run the command shown on the right. Replace [YOUR_TOKEN_HERE] with your endpoint token from cloud.vinkius.com. Use --scope user to make it available across all projects.
2

Verify the connection

Start a Claude Code session and type /mcp to list connected servers. You should see cerbos-mcp with a green status indicator.
3

Start using tools

Ask Claude Code something like "Check my latest Cerbos transactions." It will automatically discover and invoke the available Cerbos tools.

Terminal

claude mcp add --transport http cerbos-mcp https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp

Get your connection token →

Prerequisites

Claude Code CLI installed
Active Vinkius subscription with a valid endpoint token

1

Open the config file

Create or edit .mcp.json in your project root for project-level scope, or ~/.claude.json for user-level scope.
2

Add the Cerbos MCP

Paste the JSON snippet shown on the right into the mcpServers object. Replace [YOUR_TOKEN_HERE] with your endpoint token from cloud.vinkius.com.
3

Restart Claude Code

Start a new Claude Code session. Type /mcp to confirm the server is connected. The tools will be automatically available in your conversation.

.mcp.json

{
  "mcpServers": {
    "cerbos-mcp": {
      "type": "url",
      "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    }
  }
}

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Cerbos. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Cerbos now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Cerbos MCP in Claude Code

Run `claude mcp add --transport http cerbos-mcp -- ` in your terminal. Make sure all flags come before the server name. Verify the installation by running `claude mcp list`.

Yes, it uses the `authzen_evaluation` tool for single-action checks. You pass the principal and resource details, and the CLI returns the evaluation based on the AuthZEN entity model.

Your terminal agent calls `authzen_evaluations` to run multiple checks at once. It supports standard batch semantics like permit_on_first_permit, making it perfect for automated testing pipelines.

The agent executes `get_authzen_config` to return your endpoint URLs. It also uses `get_server_info` to pull version and build details directly into your terminal session.

Vinkius isolates every terminal request inside a V8 sandbox. Your principal IDs, resource conditions, and policy configurations are processed ephemerally. The zero-trust setup ensures no access data remains after the command terminates.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript