Package Information
Documentation
n8n-nodes-hudu-core
This community node enables seamless integration with the Hudu documentation platform in your n8n workflows, allowing you to automate and manage your IT documentation tasks.
What "Core" means: this is the lean, zero-runtime-dependency edition of the Hudu node — the same
HuduREST API coverage as the full n8n-nodes-hudu, minus the dedicated AI Tools node. Having zero dependencies is what lets it pass n8n's community-node verification and be installed on n8n Cloud (the hosted n8n platform at n8n.io). It talks to the same Hudu REST API whether your Hudu instance is self-hosted or Hudu-hosted. (Formerly published asn8n-nodes-hudu-cloud.)
API Compatibility: This node is aligned with Hudu API version 2.45.1 (reference:
api-docs-v2.45.1.jsonin this repository). Some features require specific API versions. Compatibility with future Hudu versions is not guaranteed without further updates.
Note — Hudu AI Tools not included in this edition: the dedicated Hudu AI Tools node (the
HuduAiToolssupplyData tool / MCP Server Trigger tool) is not part of this package. For the full feature set, including that dedicated AI Tools node, see n8n-nodes-hudu (self-hosted). This package,n8n-nodes-hudu-core, is the n8n Cloud–verifiable edition and still supports AI Agent tool use via the main Hudu node'susableAsToolcapability — every resource/operation on the Hudu node is automatically exposed to AI agents, no separate node required.
Node name & co-existence: to allow this edition and the full self-hosted n8n-nodes-hudu to be installed on the same n8n instance, this package registers a distinct node and credential. In the node picker it appears as "Hudu Core" and uses the "Hudu Core API" credential (internal types
huduCore/huduCoreApi), whereas the full package uses "Hudu" / "Hudu API". Because the node types differ, workflows built on one edition are not interchangeable with the other.
Installation
Follow the installation guide in the n8n community nodes documentation.
Credentials
To use this node, you need to:
- Have a Hudu instance
- Generate an API key in your Hudu settings
- Create a Hudu Core API credential with:
- Base URL (e.g., https://your-hudu-instance.com)
- API Key
Features
- Comprehensive pagination support for applicable resources
- Robust error handling and debugging capabilities
- Advanced filtering options with both API-side and client-side filtering
- Support for both single and bulk operations
- Dynamic loading of related resources (companies, users, assets, layouts, networks, groups, VLANs, VLAN Zones)
- Date range filtering with preset options
- Automatic type conversion and validation
- Debug logging for troubleshooting (category-based; see DEBUG.md; requires N8N_LOG_LEVEL=debug when running n8n)
- Central request sanitisation (omits empty optionals to prevent API 500s)
- Resource-specific update collections across all resources
- Optional article markdown conversion (HTML to Markdown) for Articles get/get many (uses internal regex-based converter)
- Flexible identifier support: Get operations support both numeric IDs and slugs via Identifier Type toggle (Articles, Asset Layouts, Asset Passwords, Assets, Companies, Groups, Networks, Password Folders, Procedures, Users, VLAN Zones, VLANs, Websites)
- AI Agent support: the main
Hudunode isusableAsTool, so every resource/operation can be used directly as an AI Agent tool without a separate node (compatible with Anthropic Claude, OpenAI, and other providers). For a dedicated Hudu AI Tools node with a unified per-resource tool model, seen8n-nodes-hudu(self-hosted)
Supported Resources & Operations
Activity Logs
- Get all activity logs with comprehensive filtering support
- Filter by user (ID or email), action, resource type (optionally narrowed by resource ID, filtered server-side), and date range
- Delete activity logs by datetime with optional unassigned logs filter
API Info
- Get API information and version details
Articles
- Create, update, archive, unarchive, delete, and retrieve articles
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- List all articles with comprehensive filtering (company, draft, sharing, folder, name, fuzzy search, updated date range)
- Get article version history
- Include Markdown Content toggle to return
markdown_contentalongside original HTML
Asset
- Manage core asset lifecycle & properties: Create, retrieve, update core details, archive, unarchive, move layout, and delete assets
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Link assets to companies
- Get single asset details, including all its field values
- List all assets with enhanced filtering (e.g., by company, update date, archived status)
Asset Layouts
- Create and manage asset layout templates
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Get layout fields and configurations
- List all layouts with filtering support
- Advanced custom field support: Add hints, min/max, linkable asset layout IDs, expiration, options, and more to your custom fields. Use the "Other Data" input to specify any additional API-supported properties
Asset Passwords
- Create and manage asset-related passwords
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Link passwords to assets and companies
- Filter by company and resource types
Cards
- Lookup cards by integration (supports integration ID or identifier)
- Jump to card by integration ID or identifier
- Filter by integration type and slug
Companies
- Create, update, delete, and retrieve companies
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- List all companies with filtering support
- Jump to company by integration (slug plus optional
integration_id/integration_identifier)
Expirations
- Get all expirations with comprehensive filtering (company, expiration type, resource ID/type, date ranges)
Flag Types
- Full CRUD for flag type definitions (name, color)
- Filters: name, color, slug, created_at, updated_at; supports pagination
Flags
- Full CRUD for applying flag types to records (assets, websites, articles, companies, and other supported types)
- Filters: flag_type_id, flagable_type, flagable_id, description, created_at, updated_at; supports pagination
Folders
- Create and manage document folders
- Support for nested folder structures
- Filter by parent folder
- Track child folder status
IP Addresses
- Track and manage IP addresses
- Link to companies and networks
- Filter by company and network
Lists
- Create, update, retrieve, and delete lists
- Use this resource to manage the lists themselves
- Create list items when creating a list, or add/update/remove items when updating a list
- Filter lists by name or query
List Options
- Create, update, retrieve, and delete list items within a specific list
- Use this resource to manage the items/options of a list
Magic Dash
- Create, update, get, and delete Magic Dash items
- List all items with filtering by company and title
- Delete by ID or by title + company name
Matchers
- Configure and manage integration matchers
- Filter by match status and company
- Support for sync identifiers
Networks
- Create and manage network information
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Link to companies
- Filter by company and attributes
- Pick related VLANs with an option loader
Password Folders
- Create, update, delete, and retrieve password folders
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Security modes: all users or specific groups (
allowed_groupswhensecurity = specific) - Filter by name, company, search
- Pick allowed groups with an option loader
Exports
- Initiate company exports with format options: PDF, CSV, or S3
- Optional inclusions: websites, passwords, and specific asset layouts
S3 Exports
- Initiate S3 exports (uses credentials configured in Hudu account settings)
Procedures
- Create and manage procedures
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Create from templates
- Duplicate existing procedures
- Manage procedure tasks
- Track task completion
Relations
- Create, delete and list relationships between records
- Supports all 11 API record types for both ends of a relation: Asset, Website, Procedure, AssetPassword, Company, Article, Network, IpAddress, Vlan, VlanZone, RackStorage
- Filters: created_at, description, fromable_id, fromable_type, is_inverse, toable_id, toable_type, updated_at; supports pagination
- Filtering is performed API-side and requires Hudu 2.44.1+. On older instances the API ignores these parameters and returns unfiltered results.
Groups
- Retrieve groups and retrieve a group by ID
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Filters: name, default, search; supports pagination
Label Types
- Full CRUD for label type definitions (name, color, applicable record types, access level)
- Restrict a label type to specific companies when access level is
specific_companies - Filters: name, color, slug, created_at, updated_at; supports pagination
Labels
- Full CRUD for applying label types to records (articles, assets, websites, and other supported types)
- Filters: label_type_id, labelable_type, labelable_id, user_id, created_at, updated_at; supports pagination
Users
- Get user information
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- List all users
- Filter by role and status
VLANs
- Full CRUD support for VLANs, including filtering by company, name, and VLAN ID
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Pick VLAN Zone and Status List via option loaders
VLAN Zones
- Full CRUD support for VLAN Zones, including filtering by company, name, archive status, and date ranges
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Option loader available for selecting zones where referenced
Uploads
- Manage file uploads
- Operations: Create (multipart upload), Get, Get Many, Delete
Websites
- Manage website records
- Get operation supports numeric ID or slug selection via Identifier Type toggle
- Link to companies
- Filter by company and status
- Create/update support
archivedplus monitoring flags (paused,disable_*,enable_*email security fields)
AI Agent Usage
The main Hudu node is usableAsTool, so it can be connected directly to an n8n AI Agent node — every resource and operation documented above is exposed to the agent automatically, with no separate tools node required. This works on n8n Cloud as well as self-hosted.
For a dedicated Hudu AI Tools node exposing a unified, per-resource tool model (one hudu_{resource} tool with an operation enum, structured error envelopes, describeFields/help introspection, etc.) — intended for advanced self-hosted AI Agent / MCP Server Trigger setups — see the separate n8n-nodes-hudu package (self-hosted only). That package is not submitted for n8n Cloud verification.
Resources
Development
npm run build # Compile TypeScript (n8n-node build)
npm run dev # Hot-reload development server (n8n-node dev)
npm run build:watch # TypeScript watch mode (tsc --watch)
npm run lint # Lint with n8n-node lint
npm run lint:fix # Lint and auto-fix
npm run format # Format with Prettier
npm run release # Publish via n8n-node release
Contributing
Contributions are welcome! If you'd like to contribute to this project:
- Fork the repository
- Create a feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Please ensure your PR:
- Clearly describes the changes
- Includes any relevant documentation updates
- Follows the existing code style
- Includes tests if applicable
For bug reports or feature requests, please use the GitHub issues section.
Support
If you find this node helpful and would like to support its development:
