0.7.0 • Published 5 months ago

@kaibanjs/tools v0.7.0

Weekly downloads
-
License
MIT
Repository
-
Last release
5 months ago

Kaiban Tools for AI Agents

This package provides a collection of specialized tools designed for use with AI agents, enhancing their capabilities for various tasks.

Purpose

The Kaiban Tools package offers a set of tools that can be integrated into AI agent systems, allowing agents to perform a wide range of tasks more effectively. These tools are designed to extend the capabilities of AI agents, enabling them to interact with external services, process data, and perform complex operations.

Features

  • A collection of tools specifically designed for AI agents
  • Easy integration with existing agent frameworks and architectures
  • Tools for various purposes, including web scraping, data transformation, and more
  • Configurable options for each tool to suit different agent requirements

Installation

npm install @kaibanjs/tools

Available Tools

Here's a list of all available tools. Click on the tool names to view their detailed documentation.

ToolDescriptionDocumentation
ExaAI-focused search engine using embeddings to organize web dataREADME
FirecrawlWeb scraping service for extracting structured dataREADME
GitHub IssuesGitHub API integration for fetching and analyzing repository issuesREADME
Jina URL to MDConvert web content into clean, LLM-ready markdown using Jina.aiREADME
PDF SearchExtract and search content from PDF documentsREADME
SerperGoogle Search API integration with support for multiple search typesREADME
Simple RAGBasic Retrieval-Augmented Generation implementation for Q&AREADME
Tavily SearchAI-optimized search engine for comprehensive and accurate resultsREADME
Text File SearchSearch and analyze content within text filesREADME
Website SearchSemantic search within website content using RAG modelsREADME
WolframAlphaComputational intelligence engine for complex queries and calculationsREADME
Zapier WebhookIntegration with Zapier for workflow automationREADME
Make WebhookIntegration with Make (formerly Integromat) for workflow automationREADME

Development

Local Setup

  1. Clone the repository:
git clone https://github.com/kaiban-ai/KaibanJS.git
  1. Install KaibanJS dependencies:
npm install
  1. Build the package:
npm run build
  1. Navigate to the tools package:
cd packages/tools
  1. Install dependencies:
npm install
  1. Environment Variables:

Create a .env file in the root directory with your API keys:

VITE_FIRECRAWL_API_KEY=your_firecrawl_api_key
VITE_TAVILY_API_KEY=your_tavily_api_key
VITE_SERPER_API_KEY=your_serper_api_key
VITE_EXA_API_KEY=your_exa_api_key
VITE_WOLFRAM_APP_ID=your_wolfram_app_id
  1. Run Storybook to view and test components:
npm run storybook
  1. Build the package:
npm run build
  1. Run tests:
npm run test

Contributing

To contribute a new tool:

  1. Follow the Development steps above to set up your local environment
  2. Use an existing tool as reference (check src/firecrawl or src/tavily for examples)
  3. Remember to create:
    • Your tool implementation
    • A Storybook story
    • Tests

For questions or discussions, join our Discord.

License

MIT License