Seeker-o1

Contents

• Intro
• Demo
  • video
  • Image Recognition
  • Memory
• Features & Capabilities
• Prerequisites
• Installation
• Configuration
• Quick Start
  • CLI Mode
• Usage Examples
• Basic Examples
  • Text Processing
  • Code Execution
• Contributions
• License

Intro

Seeker-o1 is a flexible open-source AI agent system. It is also an upgrade and an alternative of @Seeker the deep research agent

Demo

video

untitled.1.mp4

Image Recognition

Add the path to your image them giving a task to the agent:

task "solve this problem in the image" sample_images/deqn.png

Answer:

Memory

The agent has both short-term and long-term memory

below is a long term memory example

Seeker-o1 empowers users to create AI agents that can:

• Execute tasks through natural language instructions
• Process text inputs
• Perform basic calculations
• Run code in a controlled environment

✨ Features & Capabilities

AI Agent Architecture

• Single-Agent System: Process and execute tasks with a single agent
• Tool Integration: Use a variety of tools to accomplish tasks
• Memory Management: Basic context retention during conversation

Current Tool Ecosystem

• Text Processing:

  • Character counting
  • Word counting
  • Text transformation (uppercase, lowercase, capitalize, reverse)

• Code Execution:

  • Python code execution
  • Output capture and analysis

• Calculations:

  • Basic arithmetic operations
  • Expression evaluation

API Integration

• OpenAI API Support: Seamless integration with GPT models

Seeker-o1 supports multiple installation methods to accommodate different user preferences and environments.

Prerequisites

• Python 3.11 or higher
• pip (Python package installer)
• Git

Installation:

# Clone the repository
git clone https://github.com/iBz-04/Seeker-o1.git
cd Seeker-o1

# Create and activate a virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install in development mode
pip install -e .

Configuration

After installation, you'll need to configure Seeker-o1 with your API keys:

1. Create a .env file in the project root
2. Add your API keys:

OPENAI_API_KEY=your_openai_api_key

Quick Start

CLI Mode

Start interactive mode:

best & simplest Option

In your terminal, simply type:

seeker-o1

Alternatively :

After installing in editable or standard mode, you can launch the Seeker-o1 CLI directly:

seeker-o1 --help

This displays global options. To run a one-off task:

seeker-o1 --mode multi --task "solve the math problem in this image" assets/images/equation.png

Once inside, use help or ? to list available commands, and task to execute tasks.

📋 Usage Examples

Basic Examples

Text Processing

from seeker_o1.core.agent.tool_agent import ToolAgent

# Create an agent with text processing capabilities
agent = ToolAgent(tools=["text"])

# Process text
response = agent.execute("Count words in 'Hello, world!'")
print(response)

Code Execution

from seeker_o1.core.agent.tool_agent import ToolAgent

# Create an agent with code execution capabilities
agent = ToolAgent(tools=["code"])

# Execute Python code
response = agent.execute("Run code ```print('Hello, world!')```")
print(response)

Contributions

We welcome contributions from the community, feel free to report issues, request features or submit pull requests!

📄 License

Seeker-o1 is released under the MIT License.