---
title: "Kernel MCP integration for LuumenAI"
url: "https://www.luumen.ai/integrations/kernel-mcp/"
entity: SoftwareIntegration
category: "AI & automation"
audience: "SRE, DevOps, platform engineering, IT operations"
---

# Kernel MCP integration for LuumenAI

Drive Kernel cloud browsers, and hold the automation runs for approval

## What LuumenAI can do with Kernel MCP

LuumenAI connects to Kernel MCP with OAuth 2.0 and exposes 25 tools (3 read-only · 3 write · 19 destructive). Reads answer immediately; writes are shown as a plan and require approval by default, configurable per tool by an administrator.

## Key facts

- Authentication: OAuth 2.0
- Tools available: 25 tools (3 read-only · 3 write · 19 destructive)
- Plan: available on all plans
- Category: AI & automation

## About Kernel MCP with LuumenAI

Connect Kernel MCP and Luumen can work with your Kernel cloud-browser infrastructure from the terminal. It reads connection context, searches Kernel's documentation, and checks what the server can do before acting. Running Playwright code, shell commands, or HTTP requests inside a session is a change, so those steps are previewed and held for approval by default. Luumen signs in through OAuth with dynamic client registration.

## Example tools

Read-only (3 in total):

- `KERNEL_MCP_GET_CONNECTION_CONTEXT` — Inspect the authenticated Kernel connection before a project-scoped operation.
- `KERNEL_MCP_GET_MORE_TOOLS` — Report a genuine server capability gap only after checking the available tools and confirming none can complete the task.
- `KERNEL_MCP_SEARCH_DOCS` — Search Kernel platform documentation for guides, tutorials, and API references.

Write and destructive (22 in total, approval required by default):

- `KERNEL_MCP_BROWSER_CURL` — Send an HTTP request through an existing Kernel browser session's Chrome network stack.
- `KERNEL_MCP_COMPUTER_ACTION` — Execute computer actions on a browser session.
- `KERNEL_MCP_EXEC_COMMAND` — Execute a command synchronously inside a browser VM.
- `KERNEL_MCP_EXECUTE_PLAYWRIGHT_CODE` — Execute Playwright/TypeScript automation or browser-wide WebMCP helpers against an existing Kernel browser session.
- `KERNEL_MCP_MANAGE_API_KEYS` — Manage Kernel API keys.
- `KERNEL_MCP_MANAGE_APPS` — Manage Kernel apps when an agent needs to discover deployed app actions, invoke an app, or inspect deployment/invocation state.
- `KERNEL_MCP_MANAGE_AUTH_CONNECTIONS` — Manage reusable authenticated profiles for third-party websites.
- `KERNEL_MCP_MANAGE_BROWSER_POOLS` — Manage pre-warmed browser pools when an agent needs fast browser acquisition or reusable session capacity.

## Frequently asked questions

### How does LuumenAI connect to Kernel MCP?

Authorize once with OAuth 2.0. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.

### Can LuumenAI change things in Kernel MCP on its own?

Read actions answer immediately. Anything that writes is shown as a plan and requires approval by default, including the 19 actions classified as destructive. Administrators configure that per tool.

### Is there an audit trail?

Every call to Kernel MCP — read or write, approved or declined — is recorded with the actor, the input, and the result.

Full page: https://www.luumen.ai/integrations/kernel-mcp/