---
title: "MCP Server | Skene Cloud Docs | Skene"
description: "Connect the hosted Skene MCP server in Cursor, Claude Code, VS Code, and other MCP clients."
url: https://www.skene.ai/resources/docs/cloud/integrations/mcp
doc_version: "318d60c44b07"
last_updated: 2026-10-30
---

# MCP server

Skene MCP is a hosted MCP endpoint for coding agents. It does not replace Cursor or Claude Code. It plugs into them and gives access to workspace context, Lifecycle journey data, the tracked events inventory, analytics scans and coverage gaps, and tracking plan Check results.

Endpoint:

`https://www.skene.ai/api/mcp`

Authentication:

Pass your workspace API key in the `X-API-Key` header (see config below).

## Quick setup

### 1) Create a workspace API key

Open your workspace in Skene:

`Settings -> API keys`

Create or copy a key for MCP usage.

### 2) Add the MCP server in your client

Use this config pattern:

```json
{
  "mcpServers": {
    "skene": {
      "url": "https://www.skene.ai/api/mcp",
      "headers": {
        "X-API-Key": "your-workspace-api-key"
      }
    }
  }
}
```

### 3) Verify

Call `skene_workspace_info` from your MCP client.

If credentials and scopes are valid, the response includes workspace id, workspace slug, granted scopes, and auth method.

## Tools

These are the tools used by the workflows documented here. The available tool list depends on the server version and your credential scopes; inspect the list returned by your MCP client for the current workspace. This table is not an exhaustive registry.

| Tool | Group | Description |
| --- | --- | --- |
| `skene_workspace_info` | Connection | Returns the authenticated workspace and granted scopes for the MCP credential. |
| `skene_journey_analyse` | Lifecycle | Starts a Lifecycle rebuild and returns which schema, events, and routes to collect. |
| `skene_journey_build` | Lifecycle | Builds and saves the Lifecycle definition from the collected schema, events, and routes. |
| `skene_get_journey` | Lifecycle | Returns Lifecycle canvas data with stages, steps, levers, triggers, connectors, and usage overlays. |
| `skene_view_journey` | Lifecycle | Opens an interactive Lifecycle map for exploring stages and steps. |
| `skene_get_tracking_events` | Events | Lists tracked events in the same order as the Events page, with status, files, and mapped steps. |
| `skene_check` | Events | Starts an analytics scan and returns the scan steps for the agent to run locally. |
| `skene_gap` | Events | Compares scanned events and providers against the Lifecycle baseline and returns coverage gaps. |
| `skene_ack` | Events | Records whether flagged events are wired, deferred, or rejected, so later scans skip handled gaps. |
| `skene_analytics_review` | Events | Reviews a pull request diff for tracking regressions. It does not post to GitHub. |
| `skene_feature_check` | Plans | Checks every active tracking plan for success metric movement, goal status, and Check learning. |

## Scope and access

Tool visibility depends on scopes granted to the MCP token.

- Read tools require read scopes.
- Write tools require write scopes.
- In read-only mode, write tools are filtered out.

## Next Steps

- [Workspace API keys](https://www.skene.ai/resources/docs/cloud/workspace) - Manage API keys and access
- [GitHub integration](https://www.skene.ai/resources/docs/cloud/integrations/github) - PR reviews and tracking checks

## Sitemap

- [Markdown sitemap](https://www.skene.ai/sitemap.md): every page on this site, with its markdown counterpart.
- [Index for language models](https://www.skene.ai/llms.txt): the same set as a link-dense index.
- [Full corpus](https://www.skene.ai/llms-full.txt): reference content concatenated as plain text.
