vectorvault

VectorVault

VectorVault API - JavaScript Client: Streamline your front-end development with the powerful capabilities of VectorVault's Cloud Vector Database. This JavaScript client provides seamless integration for building advanced RAG (Retrieve and Generate) applications. Whether you're working with JavaScript, HTML, or other web technologies, our API simplifies the process of fetching RAG responses through API POST requests. This package is the key to unlocking quick and efficient development for AI-powered web applications, ensuring a secure and robust connection to the VectorVault ecosystem. Start crafting exceptional RAG apps with minimal effort and maximum efficiency.

Installation

1. Via NPM (for React/Node.js/bundled projects)

npm install vectorvault --save

2. Via CDN (for HTML)

<script src="https://cdn.jsdelivr.net/gh/John-Rood/VectorVault-js@main/dist/vectorvault.bundle.js"></script>

Usage

To use VectorVault, you need to import it, instantiate it, and log in with your user credentials:

import VectorVault from 'vectorvault';

// Create a VectorVault instance 
const vectorVault = new VectorVault(); 

// Login with password
await vectorVault.login('[e-mail]', 'your_password');

// OR Login with api
await vectorVault.loginAPI('[e-mail]', 'your_api_key'); 

Basic Operations

After logging in, you can perform various operations with your vaults.

Get a Chat Response

const params = {
  vault: 'your_vault_name',
  text: 'Your query here',
  // ...other optional parameters
};

vectorVault
  .getChat(params)
  .then(response => console.log(response))
  .catch(error => console.error(error));

Manage Items in the Vault

Get Items

const vault = 'your_vault_name';
const itemIds = [1, 2, 3]; // IDs of the items you want to retrieve

vectorVault
  .getItems(vault, itemIds)
  .then(items => console.log(items))
  .catch(error => console.error(error));

Add New Items to the Cloud

const params = {
  vault: 'your_vault_name',
  text: 'Your text data here',
  // ...other optional parameters
};

vectorVault
  .addCloud(params)
  .then(response => console.log(response))
  .catch(error => console.error(error));

Streaming Chat Responses

The getChatStream function allows you to stream data from the VectorVault API. It requires two arguments: params, which is an object containing the parameters for your request, and callback, which is a function that will be called with each piece of data received from the stream.

Here's how you can use getChatStream:

// Define your callback function
function handleStreamedData(data) {
  // Process the streamed data here
  console.log(data);
}

// Set up the parameters for your request
const streamParams = {
  vault: 'your_vault_name',
  text: 'Your query here',
  // ...other optional parameters
};

// Start streaming data
vectorVault
  .getChatStream(streamParams, handleStreamedData)
  .then(() => console.log('Streaming completed.'))
  .catch(error => console.error('Streaming error:', error));

Parameters for getChatStream

The params object can include any of the following properties:

• vault: The name of your vault.
• text: The input text for the chat.
• history: The chat history, if applicable.
• summary: A boolean indicating if the response should be a summary.
• get_context: A boolean to indicate if you want to receive context information.
• n_context: The number of context turns you want to receive.
• return_context: A boolean to include the context in the response.
• smart_history_search: A boolean to enable smart history searching.
• model: The model you want to use, e.g., "gpt-3.5-turbo".
• include_context_meta: A boolean to include metadata about the context.
• metatag, metatag_prefixes, metatag_suffixes: Arrays for advanced context tagging.
• custom_prompt: A custom prompt to be used instead of the default.
• temperature: The creativity temperature.
• timeout: The timeout for the model response wait time.

Make sure to replace "Your query here" with the actual text you want to send to the API.

Note: getChatStream is an asynchronous function and should be handled with async/await or .then().catch() for proper error handling.

Additional Operations

Authentication

• Login

  await vectorVault.login('[e-mail]', 'your_password');
  // OR 
  await vectorVault.loginAPI('[e-mail]', 'your_api_key');
  

• Logout

  vectorVault.logout();
  

Vault Management

• Get Vaults

  vectorVault
    .getVaults()
    .then(vaults => console.log(vaults))
    .catch(error => console.error(error));
  

• Delete Vault

  const vault = 'your_vault_name';
  vectorVault
    .deleteVault(vault)
    .then(response => console.log(response))
    .catch(error => console.error(error));
  

Data Management

• Edit an Item

  const vault = 'your_vault_name';
  const itemId = 123;
  const newText = 'Updated text content';
  
  vectorVault
    .editItem(vault, itemId, newText)
    .then(response => console.log(response))
    .catch(error => console.error(error));
  

• Delete Items

  const vault = 'your_vault_name';
  const itemIds = [1, 2, 3];
  
  vectorVault
    .deleteItems(vault, itemIds)
    .then(response => console.log(response))
    .catch(error => console.error(error));
  

• Get Total Items

  const vault = 'your_vault_name';
  
  vectorVault
    .getTotalItems(vault)
    .then(total => console.log(`Total items: ${total}`))
    .catch(error => console.error(error));
  

Similarity and Distance

• Get Distance Between Two Items

  const vault = 'your_vault_name';
  const id1 = 1;
  const id2 = 2;
  
  vectorVault
    .getDistance(vault, id1, id2)
    .then(distance => console.log(`Distance: ${distance}`))
    .catch(error => console.error(error));
  

• Get Similar Items

  const params = {
    vault: 'your_vault_name',
    text: 'Sample text to find similarities',
    num_items: 4,
    include_distances: true,
    // ...other optional parameters
  };
  
  vectorVault
    .getSimilar(params)
    .then(similarItems => console.log(similarItems))
    .catch(error => console.error(error));
  

Account and Vault Data

• Get Account Data

  vectorVault
    .getAccountData()
    .then(data => console.log(data))
    .catch(error => console.error(error));
  

• Download Vault Data to JSON

  const params = {
    vault: 'your_vault_name',
    return_meta: true,
  };
  
  vectorVault
    .downloadToJson(params)
    .then(jsonData => console.log(jsonData))
    .catch(error => console.error(error));
  

• Upload Data from JSON

  const vault = 'your_vault_name';
  const jsonData = {/* Your JSON data */};
  
  vectorVault
    .uploadFromJson(vault, jsonData)
    .then(response => console.log(response))
    .catch(error => console.error(error));
  

Customization

• Save Personality Message

  const vault = 'your_vault_name';
  const personalityMessage = 'Your personality message here';
  
  vectorVault
    .savePersonalityMessage(vault, personalityMessage)
    .then(response => console.log(response))
    .catch(error => console.error(error));
  

• Fetch Personality Message

  const vault = 'your_vault_name';
  
  vectorVault
    .fetchPersonalityMessage(vault)
    .then(message => console.log(message))
    .catch(error => console.error(error));
  

• Save Custom Prompt

  const vault = 'your_vault_name';
  const customPrompt = 'Your custom prompt here';
  
  vectorVault
    .saveCustomPrompt(vault, customPrompt)
    .then(response => console.log(response))
    .catch(error => console.error(error));
  

• Fetch Custom Prompt

  const vault = 'your_vault_name';
  
  vectorVault
    .fetchCustomPrompt(vault)
    .then(prompt => console.log(prompt))
    .catch(error => console.error(error));
  

3D Map Data

• Fetch 3D Map Data

  const vault = 'your_vault_name';
  const highlightId = null; // or specify an item ID to highlight
  
  vectorVault
    .fetch3DMap(vault, highlightId)
    .then(mapData => console.log(mapData))
    .catch(error => console.error(error));
  

Flow Execution with Streaming

• Run Flow with Streaming Response

runFlow(                 // returns full response after completion
runFlowStream(           // streams logs & messages
    flowName,            // string – Flow identifier
    message,             // string – user message
    history = '',        // string – chat history (optional)
    conversation_user_id = null, // any – used by multi‑user apps
    session_id = null,   // any – thread identifier
    invoke_method = null,// string – arbitrary label written to logs
    internal_vars = null,// object – {varName: value} overrides flow‑internal vars
    callbacks = {        // only used by runFlowStream
        onLog: log => {},      // fires for every 'log' event
        onMessage: chunk => {},// fires for every 'message' chunk
        onError: err => {}     // fires on network / parsing errors
    }
)

const flow = 'my_flow';
const msg  = 'Hi, Flow!';
const payload = { role: 'tester' };
const cb = {
  onLog:   log => console.log('[LOG]', log),
  onMessage: m => process.stdout.write(m),
  onError:  err => console.error(err)
};

vectorVault
  .runFlowStream(flow, msg, '', null, null, 'CLI', payload, cb)
  .then(res => console.log('\nFlow done:', res))
  .catch(err => console.error('Flow error:', err));

Error Handling

Most methods return promises and should be handled with .then().catch() or async/await syntax. Proper error handling ensures your application can handle exceptions gracefully.

Notes

• Replace placeholder values (like '[e-mail]', 'your_password', 'your_vault_name', etc.) with your actual account and vault information.
• If you don't already have a VectorVault account, sign up at vectorvault.io.

Conclusion

With the VectorVault JavaScript client, you can easily integrate advanced vector-based operations into your web applications. Whether you're building AI Agents, recommendation systems, or any AI-powered application, VectorVault provides the tools you need for efficient and secure development.