Skip to main content

AI Capabilities in HAP

HAP offers a range of built-in AI features that enhance automation, data entry, and user experience across your application:

  • ChatBot
  • Vector Knowledge Base
  • AI Worksheet Creation
  • AI Custom Fields
  • AI Sample Data Generation
  • AI Record Creation
  • AI Custom Field
  • AI Agent
  • AI Action
  • AI Code Block Generation
  • AI Knowledge Retrieval
  • AI Generate Text/Data Object
  • App Multilingual Support
  • AI Configuration for HAP Server
  • Extensions
    • HAP for VS Code
    • Dify

ChatBot​

Users can interact with the ChatBot via multi-turn conversations. The ChatBot can identify user intent, perform create/update/delete operations on app data, summarize or analyze content, and drive business automation.

Vector Knowledge Base​

The Vector Knowledge Base is HAP’s built-in semantic retrieval infrastructure. By vectorizing worksheet records, attachments, and discussions, it enables AI Agents and workflows to retrieve relevant knowledge based on semantic understanding rather than simple keyword matching.

AI Worksheet Creation​

Based on user descriptions, AI can automatically create worksheets, including default icons, fields, and layout configurations.

AI Sample Data Generation​

AI can generate sample records within worksheets, making it easier for builders to test and demonstrate their applications.

AI Record Creation​

AI intelligently parses user input or uploaded images and fills the data into the corresponding fields in a worksheet.

AI Custom Fields​

You can generate fully customized field controls using AI-generated code via conversation. These fields support tailored UI and interaction logic. If you're familiar with React and Tailwind CSS, you can write or modify the code directly for further customization. Learn more about custom fields

AI Agent​

The AI Agent node in workflows can analyze user input, perform reasoning, and execute tools such as data querying, record creation, updates, API calls, or packaged business processes.

AI Action​

AI Action provides an entry point to start record-based conversations directly from a record detail page.

Click any AI Action to open a multi-turn conversation centered on the current record. The system automatically injects contextual data—such as record fields, discussions, and logs—and continuously generates results based on the configured prompt, model, and tools.

AI-Generate Text/Data Object​

Leverage AIGC capabilities to assist with content creation and improve productivity. This includes the AI Text Generation and AI Data Object Generation workflow nodes.

Knowledge Retrieval Node​

The Knowledge Retrieval node in workflow supports multi-modal retrieval (semantic search, keyword search, and hybrid search) across multiple knowledge bases and applications. It returns highly relevant results as structured arrays, including text snippets, sources, and relevance scores, which can be used for AI Q&A, data population, or workflow automation.

AI Code Block Generation​

In the Code Block Node within a workflow, AI can generate JavaScript or Python code based on a functional description provided by the user. The generated code complies with block specifications and is ready to use.

App Multilingual Support​

Powered by advanced natural language processing models, the AI Translation feature offers one-click translation of your app's content. This allows your application to easily support multilingual collaboration scenarios.

AI Configuration in HAP Server​

To support AI capabilities in HAP Server, HAP offers two configuration options:

  1. Configure your own LLM API key
  2. Deploy your own LLM service

MCP​

MCP enables AI-assisted development. Generate an MCP Server configuration using your API key, and add the JSON configuration below to your IDE’s MCP configuration file. When API documentation is updated, the MCP Server will automatically stay in sync.

Extensions​

HAP for VS Code​

HAP for VS Code is the official VS Code extension for HAP. It integrates the entire development lifecycle—including building, debugging, submitting, and publishing view plugins—into a single sidebar panel, allowing you to complete all tasks without leaving the editor.

Dify​

In Dify, you can use HAP’s built-in tools to create an application assistant that enables Q&A based on real-time application data.

Was this document helpful?