Metadata-Version: 2.4
Name: langchain-anakin
Version: 0.1.0
Summary: LangChain integration for Anakin API - web scraping, AI search, and intelligent data extraction
Author-email: Anakin AI <anakin@anakininc.com>
License: MIT
Project-URL: Homepage, https://anakin.io
Project-URL: Documentation, https://anakin.io/docs
Project-URL: Repository, https://github.com/Anakin-Inc/langchain-anakin
Project-URL: Bug Tracker, https://github.com/Anakin-Inc/langchain-anakin/issues
Keywords: langchain,anakin,web scraping,ai search,perplexity,data extraction,agentic search,tools
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Topic :: Internet :: WWW/HTTP
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: langchain-core>=0.1.0
Requires-Dist: requests>=2.31.0
Requires-Dist: pydantic>=2.0.0
Provides-Extra: dev
Requires-Dist: pytest>=7.0.0; extra == "dev"
Requires-Dist: pytest-asyncio>=0.21.0; extra == "dev"
Requires-Dist: black>=23.0.0; extra == "dev"
Requires-Dist: mypy>=1.0.0; extra == "dev"
Requires-Dist: ruff>=0.1.0; extra == "dev"
Dynamic: license-file

# langchain-anakin

LangChain integration for [Anakin AI](https://anakin.io) - powerful web scraping, AI-powered search, and intelligent data extraction tools for your LangChain applications.

[![PyPI version](https://badge.fury.io/py/langchain-anakin.svg)](https://badge.fury.io/py/langchain-anakin)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

## Features

- 🔧 **Three Powerful LangChain Tools**:
  - `AnakinScrapeTool` - Scrape websites and extract structured data
  - `AnakinSearchTool` - AI-powered search with Perplexity AI
  - `AnakinAgenticSearchTool` - Advanced multi-stage AI pipeline for comprehensive research
- 🔐 Simple authentication with API key
- 📦 Easy integration with LangChain agents and chains
- ⚡ Asynchronous operations with automatic polling
- 🎯 Type-safe with Pydantic models
- 📖 Comprehensive documentation and examples

## Installation

```bash
pip install langchain-anakin
```

## Quick Start

### 1. Get Your API Key

Sign up at [Anakin.io](https://anakin.io) to get your API key.

### 2. Use the Tools

```python
from langchain_anakin import AnakinScrapeTool, AnakinSearchTool, AnakinAgenticSearchTool

# Initialize tools with your API key
api_key = "your-anakin-api-key"

scrape_tool = AnakinScrapeTool(api_key=api_key)
search_tool = AnakinSearchTool(api_key=api_key)
agentic_tool = AnakinAgenticSearchTool(api_key=api_key)

# Use with LangChain agents
from langchain.agents import initialize_agent, AgentType
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(temperature=0)

tools = [scrape_tool, search_tool, agentic_tool]

agent = initialize_agent(
    tools=tools,
    llm=llm,
    agent=AgentType.OPENAI_FUNCTIONS,
    verbose=True
)

# Ask the agent to use the tools
result = agent.run(
    "Search for the latest AI trends and scrape the top 3 sources"
)
```

## Tools Overview

### 1. AnakinScrapeTool

Scrape websites and extract content including HTML, markdown, and structured data.

```python
from langchain_anakin import AnakinScrapeTool

tool = AnakinScrapeTool(api_key="your-api-key")

# Direct invocation
result = tool.invoke({
    "url": "https://example.com/product",
    "country": "us",
    "force_fresh": False
})

print(result)
```

**Input Parameters:**
- `url` (required): Website URL to scrape
- `country` (optional): Proxy country code (default: "us")
- `force_fresh` (optional): Bypass cache (default: False)
- `max_wait_time` (optional): Max seconds to wait (default: 300)
- `poll_interval` (optional): Seconds between checks (default: 3)

**Use Cases:**
- Extract product information from e-commerce sites
- Scrape article content for analysis
- Monitor website changes
- Gather structured data from web pages

### 2. AnakinSearchTool

Perform AI-powered searches with instant answers and citations.

```python
from langchain_anakin import AnakinSearchTool

tool = AnakinSearchTool(api_key="your-api-key")

# Direct invocation
result = tool.invoke({
    "query": "What are the latest developments in quantum computing?",
    "max_results": 5
})

print(result)
```

**Input Parameters:**
- `query` (required): Search question or query
- `max_results` (optional): Maximum results to return (default: 5)

**Use Cases:**
- Research and fact-checking
- Competitive intelligence
- Real-time information gathering
- Content research
- Answer complex questions

### 3. AnakinAgenticSearchTool

Advanced multi-stage AI pipeline that automatically searches, scrapes, and extracts structured data.

```python
from langchain_anakin import AnakinAgenticSearchTool

tool = AnakinAgenticSearchTool(api_key="your-api-key")

# Direct invocation
result = tool.invoke({
    "prompt": "Find pricing information for top 5 project management tools",
    "use_browser": True
})

print(result)
```

**Input Parameters:**
- `prompt` (required): Research prompt or question
- `use_browser` (optional): Use browser for scraping (default: True)
- `max_wait_time` (optional): Max seconds to wait (default: 600)
- `poll_interval` (optional): Seconds between checks (default: 5)

**Use Cases:**
- Comprehensive market research
- Competitive analysis with structured data
- Lead generation with enrichment
- Automated data collection for reports
- Multi-source information synthesis

## Advanced Examples

### Example 1: Web Research Agent

```python
from langchain_anakin import AnakinSearchTool, AnakinScrapeTool
from langchain_openai import ChatOpenAI
from langchain.agents import initialize_agent, AgentType

api_key = "your-anakin-api-key"

# Initialize tools
search_tool = AnakinSearchTool(api_key=api_key)
scrape_tool = AnakinScrapeTool(api_key=api_key)

# Create agent
llm = ChatOpenAI(model="gpt-4", temperature=0)
agent = initialize_agent(
    tools=[search_tool, scrape_tool],
    llm=llm,
    agent=AgentType.OPENAI_FUNCTIONS,
    verbose=True
)

# Run research task
result = agent.run(
    "Research the top 3 AI companies in 2026 and provide detailed information about each"
)
```

### Example 2: Market Research with Agentic Search

```python
from langchain_anakin import AnakinAgenticSearchTool
from langchain_openai import ChatOpenAI
from langchain.agents import initialize_agent, AgentType

# Initialize tool
agentic_tool = AnakinAgenticSearchTool(api_key="your-anakin-api-key")

# Create specialized agent
llm = ChatOpenAI(model="gpt-4", temperature=0)
agent = initialize_agent(
    tools=[agentic_tool],
    llm=llm,
    agent=AgentType.OPENAI_FUNCTIONS,
    verbose=True
)

# Perform comprehensive research
result = agent.run(
    "Compare pricing, features, and user reviews of Salesforce, HubSpot, and Zoho CRM"
)
```

### Example 3: Custom Chain with Multiple Tools

```python
from langchain_anakin import (
    AnakinScrapeTool,
    AnakinSearchTool,
    AnakinAgenticSearchTool
)
from langchain.chains import LLMChain
from langchain_openai import ChatOpenAI
from langchain.prompts import PromptTemplate

# Initialize all tools
api_key = "your-anakin-api-key"
scrape = AnakinScrapeTool(api_key=api_key)
search = AnakinSearchTool(api_key=api_key)
agentic = AnakinAgenticSearchTool(api_key=api_key)

# Use tools in a custom workflow
# 1. Search for topics
search_result = search.invoke({
    "query": "Best AI tools for developers in 2026"
})

# 2. Extract URLs from search and scrape them
# (In a real scenario, you'd parse URLs from search_result)
scrape_result = scrape.invoke({
    "url": "https://example.com/ai-tools"
})

# 3. Do comprehensive analysis
agentic_result = agentic.invoke({
    "prompt": "Analyze features and pricing of the AI tools mentioned"
})
```

## Configuration

### Custom Base URL

If you're using a self-hosted or custom Anakin API endpoint:

```python
from langchain_anakin import AnakinScrapeTool

tool = AnakinScrapeTool(
    api_key="your-api-key",
    base_url="https://your-custom-api.com"
)
```

### Tool Timeouts and Polling

Adjust timeouts and polling intervals for long-running operations:

```python
# For scraping with custom timeout
result = scrape_tool.invoke({
    "url": "https://example.com",
    "max_wait_time": 600,  # 10 minutes
    "poll_interval": 5  # Check every 5 seconds
})

# For agentic search with custom settings
result = agentic_tool.invoke({
    "prompt": "Research prompt",
    "max_wait_time": 900,  # 15 minutes
    "poll_interval": 10  # Check every 10 seconds
})
```

## Error Handling

All tools handle errors gracefully and return descriptive error messages:

```python
from langchain_anakin import AnakinScrapeTool

tool = AnakinScrapeTool(api_key="your-api-key")

try:
    result = tool.invoke({"url": "https://invalid-url"})
except Exception as e:
    print(f"Error: {e}")
```

## API Reference

### AnakinScrapeTool

```python
class AnakinScrapeTool(BaseTool):
    """Tool for scraping websites using the Anakin API."""
    
    name: str = "anakin_scrape_url"
    api_key: str  # Required
    base_url: str = "https://api.anakin.io"  # Optional
```

### AnakinSearchTool

```python
class AnakinSearchTool(BaseTool):
    """Tool for AI-powered searches using the Anakin API."""
    
    name: str = "anakin_ai_search"
    api_key: str  # Required
    base_url: str = "https://api.anakin.io"  # Optional
```

### AnakinAgenticSearchTool

```python
class AnakinAgenticSearchTool(BaseTool):
    """Tool for advanced agentic searches using the Anakin API."""
    
    name: str = "anakin_agentic_search"
    api_key: str  # Required
    base_url: str = "https://api.anakin.io"  # Optional
```

## Development

### Setup Development Environment

```bash
# Clone the repository
git clone https://github.com/Anakin-Inc/langchain-anakin.git
cd langchain-anakin

# Install dependencies
pip install -r requirements-dev.txt

# Install package in editable mode
pip install -e .
```

### Running Tests

```bash
pytest
```

### Code Formatting

```bash
black langchain_anakin/
ruff check langchain_anakin/
```

## Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (`git checkout -b feature/AmazingFeature`)
3. Commit your changes (`git commit -m 'Add some AmazingFeature'`)
4. Push to the branch (`git push origin feature/AmazingFeature`)
5. Open a Pull Request

## Support

- 📖 [API Documentation](https://anakin.io/docs)
- 💬 [Community Forum](https://community.anakin.io)
- 🐛 [Report Issues](https://github.com/Anakin-Inc/langchain-anakin/issues)
- 📧 Email: anakin@anakininc.com

## License

MIT License - see the [LICENSE](LICENSE) file for details.

## Related Projects

- [n8n-nodes-anakin-scraper](https://github.com/Anakin-Inc/n8n-nodes-anakin-scraper) - n8n integration
- [Anakin API](https://anakin.io) - Official API documentation

## Changelog

### 0.1.0 (2026-01-23)

- Initial release
- Added `AnakinScrapeTool` for web scraping
- Added `AnakinSearchTool` for AI-powered search
- Added `AnakinAgenticSearchTool` for agentic research
- Full LangChain integration with BaseTool
- Comprehensive documentation and examples
