durin
Sign up

Give AI access. Keep control.

Control AI access through Durin with policies, optional approvals and audit trails.

Sign upTry without an account
▣ ✻ Claude Code

Try an MCP request.

See what an agent is allowed to do. Run a sample MCP request through Durin’s policy checks without an account or API key.

Claude Code v2.1.278

Fable 5.1 with high effort · Claude Max

/Users/durin

⚠ 1 MCP server needs authentication · run /mcp

  1. YouPrompt

    Create the roadmap ticket now that approval demo-approval-001 is present.

  2. ClaudeThinking

    The write still needs the same arguments and matching approval. I will send both through Durin before execution.

  3. Tool useMCP call

    mcp__durin__sandbox__create_ticket({ "resource": "engineering/roadmap", "title": "Evaluate governed agent access", "durinIdempotencyKey": "demo-write-001", "durinApprovalId": "demo-approval-001" })

  4. Durin MCPPolicy check

    ALLOW: Approval matched. A ticket response is returned.

  5. Tool resultSandbox response

    Synthetic ticket DEMO-001: Evaluate governed agent access. No real ticket was created.

  6. ClaudeFinal answer

    Created sample ticket DEMO-001 in the sandbox response.

Press Enter to run the prefilled command. If you do nothing, it plays automatically.

Install Durin to your Agent

Copy a setup prompt into your Agent, or choose your client for manual setup.

Let your Agent set it up

Copy this prompt into your Agent. It runs the Durin CLI and guides setup. You choose your organization and Agent in the CLI and complete sign-in in the browser.

Manual setup

Choose your client and configure a direct MCP connection.

Use your personal endpoint

Copy your endpoint from Connected clients in Durin. Replace YOUR_PERSONAL_ENDPOINT_ID in the example with the value from your endpoint.

Get your MCP endpoint

Run in your terminal:

Open Claude Code and use /mcp to authenticate Durin.

Then request a tool through Durin and confirm the session in Connected clients.

Configuration for Claude Code

~/.claude.json

Merge the Durin entry into your existing configuration.

Move from shadow connections to a managed path.

Durin provides MCP Governance for teams that need a clear route from an approved agent client to an approved company system. Evaluate the boundary, then plan the rollout around the coverage you can verify.

  1. A route you can inspect

    Employees can use approved agent clients through a managed route to approved connections, tools, and resources. A client login is not treated as permission to every upstream system.

  2. Decide before the call

    Requests meet the current membership, connection, resource, and policy boundary. Missing or unsupported facts fail closed. Sensitive writes can require an independent approver with an exact, time-limited grant.

  3. Keep the evidence

    Durin records the requesting person or service, client, connection, policy decision, approval where required, and execution result. Potentially completed writes remain explicitly uncertain instead of being blindly retried.

Connections, not people.

First connection free. Your first four paid connections are $399 each per month. Each additional paid connection is $199 per month. 10 GB per connection monthly, pooled across your organization and prorated by enabled time. People and approved agent clients do not add seat charges.

See how pricing works

Make the case for governed AI.

Your next shadow IT problem is an AI agent.

A practical guide for IT teams evaluating MCP Governance and a managed route for AI agent access.

A trusted employee. An overprivileged agent.

A CISO guide to unmanaged agent access, MCP Governance boundaries, prompt injection, and production gates.

MCP Governance architecture for CTOs.

A CTO guide to MCP Governance architecture, trust boundaries, approval semantics, and production trust.

When AI can act, who is accountable?

A CEO guide to governing AI agent access, reducing shadow connections, and setting responsible rollout gates.

What is MCP Governance, and when do you need it?

Learn where MCP Governance fits, which access controls to evaluate, and when a shared tool connection needs centralized governance.

MCP Governance vs API gateway: which controls do you need?

Compare MCP Governance with API gateway controls, check what your existing platform supports, and identify gaps in tool policy, approvals, and audit.

AI agent approval workflows: review the action before it runs

Design human review for sensitive tool calls, bind approval to the exact request, and handle retries without confusing permission with execution.

MCP Governance: a practical rollout checklist

Inventory agent access, review tools and credentials, configure policy, test approvals, and keep useful evidence as your MCP rollout grows.

Centralize MCP access with Durin

See how Durin connects AI clients to approved tools, checks access, handles write approvals, and records decisions through a governed MCP path.

See your governed access path.

Create an account to manage identity, policy, approvals, and audit for requests routed through Durin.

Create an account