WS.Eng MCP Server

The MCP server for the WS.Eng CLI and Data API. Exposes cross-cutting engineering tools, prompts, and resources.

Version: 1.0.0

Connect to WS.Eng MCP Server

Local (CLI)

Add the following to your ~/.cursor/mcp.json file:

{
  "mcpServers": {
    "ws-eng-mcp-server": {
      "command": "wseng",
      "args": [
        "mcp"
      ]
    }
  }
}

Remote (HTTP)

Add the following to your ~/.cursor/mcp.json file:

{
  "mcpServers": {
    "ws-eng-mcp-server": {
      "type": "http",
      "url": "/mcp",
      "headers": {
        "X-Api-Key": "<API_KEY>"
      }
    }
  }
}

Local (CLI)

Add the following to your VS Code settings.json:

{
  "mcp.servers": {
    "ws-eng-mcp-server": {
      "type": "stdio",
      "command": "wseng",
      "args": [
        "mcp"
      ]
    }
  }
}

Remote (HTTP)

Add the following to your VS Code settings.json:

{
  "mcp.servers": {
    "ws-eng-mcp-server": {
      "type": "http",
      "url": "/mcp",
      "headers": {
        "X-Api-Key": "<API_KEY>"
      }
    }
  }
}

Local (CLI)

Run the following command:

claude mcp add --transport stdio ws-eng-mcp-server -- wseng mcp

Remote (HTTP)

Run the following command:

claude mcp add --transport http ws-eng-mcp-server /mcp --header "X-Api-Key: <API_KEY>"

Tools

The server exposes the following MCP tools:

Tool	Description
`wseng_get_data_api_metadata`	Get Data API documentation. The Data API exposes data about the engineers, repositories, products, tickets, and GitHub Actions workflows+logs from our engineering organization. By default, returns a compact list of all endpoints with operationId, path, and description. Use operationIds to get full details (parameters, response schema) for specific endpoints. Use full=true to get complete documentation for all endpoints.
`wseng_query_data_api`	Query a specific Data API endpoint. Use wseng_get_data_api_metadata first to discover available endpoints.
`wseng_review_work`	Review the work done for a ticket.
`wseng_list_documents`	List all available shared documents. This includes documents of several types: context packs (documents that are scoped to a specific product or subsystem and summarize the functionality, main decisions, business context, architecture, etc.), second brains/brain lifts (documents that encapsulate our principles, best practices, guidelines and approaches) and quality bars (lower level rules that enforce specific practices within our code).
`wseng_get_document`	Get the content of a shared document by name or URI. Use wseng_list_documents first to discover available documents.
`wseng_web_search`	Search the web using multiple providers. Supports three modes: 'search' returns raw results from Perplexity, Brave, Tavily, Exa, and SERP; 'ask' (default) summarizes combined results into a comprehensive answer; 'research' performs deep research using Perplexity and OpenAI deep research models.
`wseng_generate_changelog`	Generate a changelog and PR overview for a ticket. Fetches PR diffs, runs LLM generation, and writes to the ticket output. Workflow: 1) Resolves ticket and fetches PR diffs. 2) Generates changelog + summary via LLM. 3) Writes result to ticket output under the 'changelog' section. Use when the user asks to: 'generate the changelog for WEW-123', 'fill in the output for this ticket', 'create a changelog'. For regeneration with changes: pass the feedback param. Prerequisites: ticket must exist and have linked PRs (or provide prUrls). If no PRs found, use wseng_query_data_api to check the ticket.
`wseng_generate_how_to_test`	Generate how-to-test instructions for a ticket. Not yet implemented — coming in a future update.
`wseng_generate_story_output`	Generate the full story output for a ticket. Not yet implemented — coming in a future update.
`wseng_update_ticket_output`	Manually update a specific section of a ticket's output. Not yet implemented — coming in a future update.

Configuration