HUDU

You run an MCP server that connects HUDU data capabilities to your language model workflows. This server exposes company, knowledge base, assets, users, networks, procedures, and activity data from HUDU, enabling you to search, retrieve, and filter information efficiently as you build assistant experiences.

How to use

Your HUDU MCP Server lets an MCP client query HUDU data sources through a secure, typed interface. You can search for companies, articles, assets, users, networks, and procedures, then drill into details or listings with advanced filters. Use the available tools to retrieve lists, fetch details by ID, or access filtered results. When you request large datasets, you’ll see pagination and truncation behavior so you can fetch data in manageable chunks.

How to install

Prerequisites you need before running the MCP server:

• Node.js 18.0.0 or higher.

• A HUDU instance with API access and a HUDU API key.

Step-by-step commands to set up and run the server locally:

Clone the MCP server repository, install dependencies, and build:

git clone https://github.com/jlbyh2o/hudu-mcp.git
cd hudu-mcp
npm install
npm run build

Additional sections

Configuration, security, and operational notes help you run the MCP server reliably and securely.

Environment variables you must set in your runtime environment:

• HUDU_API_KEY: Your HUDU API key.

• HUDU_BASE_URL: Your HUDU instance domain, without any /api/v1 path.

Startup commands you can use to run the server in development or production mode:

# Development mode
npm run dev

# Production mode
npm start

Notes on usage and troubleshooting

During startup, the MCP server validates your HUDU API connection by querying the API info endpoint. This ensures the API key is valid, the base URL is reachable, and the key has the necessary permissions. If validation fails, you’ll receive clear errors indicating invalid keys, insufficient permissions, or connectivity problems.

Large responses are managed automatically to avoid token size limits. If a response is too large, you’ll see truncation indicators and you should use pagination to fetch data in smaller chunks.

If you work with Claude Desktop or another MCP client, you can configure the client to run the server locally as a standard process and supply the required environment variables to your session.

Security and best practices

• API keys are passed via environment variables and are protected by the MCP runtime.

• Password data is masked in list responses for security.

• All API requests include authentication headers and input is validated using strict schemas.

Troubleshooting

Common issues include configuration validation failures, authentication errors, and network connectivity problems. Check that HUDU_API_KEY and HUDU_BASE_URL are set correctly and that the HUDU instance is accessible from your environment. If you encounter 401 or 403 errors, verify that your API key is active and has the required permissions.

For large datasets, ensure you use page and page_size parameters to fetch data in smaller chunks and look for truncation indicators in responses to adjust your query strategy.

Available tools

get_companies

Retrieve a list of companies/customers with advanced filtering such as name, phone, website, city, state, slug, and more; supports pagination and search terms.

get_company_details

Get detailed information about a specific company using its ID.

search_articles

Search knowledge base articles with filters like company_id, name, slug, draft status, sharing, updated_at, and pagination.

get_article

Fetch detailed content for a specific knowledge base article by ID.

get_assets

Retrieve assets with advanced filtering options including company_id, asset_layout_id, serials, archival status, and pagination.

get_asset_passwords

Retrieve password assets (credentials) for a company with optional filtering and pagination.

get_asset_layouts

Retrieve asset layout definitions with optional filtering and pagination.

get_asset_layout

Get detailed information about a specific asset layout by ID.

get_users

Retrieve a list of users with optional search and pagination.

get_user

Get detailed information about a specific user by ID.

get_networks

Retrieve a list of networks with optional search and pagination.

get_network

Get detailed information about a specific network by ID.

get_procedures

Retrieve a list of procedures with optional filtering and pagination.

get_procedure

Get detailed information about a specific procedure by ID.

get_activity_logs

Retrieve activity logs with advanced filtering by user, resource, action, date, and pagination.

get_folders

Retrieve a list of folders with optional search and pagination.

get_folder

Get detailed information about a specific folder by ID.