https://github.com/namikazi25/ecobot
EcoBot is a chatbot built to empower ecological exploration and species identification. It leverages GPT-4o Mini for intelligent responses, supports chat history persistence, and can analyze images and PDFs to provide insights into biodiversity, species traits, and ecological questions.
https://github.com/namikazi25/ecobot
chatbot ecology langchain python
Last synced: about 2 months ago
JSON representation
EcoBot is a chatbot built to empower ecological exploration and species identification. It leverages GPT-4o Mini for intelligent responses, supports chat history persistence, and can analyze images and PDFs to provide insights into biodiversity, species traits, and ecological questions.
- Host: GitHub
- URL: https://github.com/namikazi25/ecobot
- Owner: namikazi25
- Created: 2025-01-19T23:55:16.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2025-02-23T23:14:41.000Z (over 1 year ago)
- Last Synced: 2025-02-23T23:26:39.421Z (over 1 year ago)
- Topics: chatbot, ecology, langchain, python
- Language: Python
- Homepage:
- Size: 915 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 🌿 EcoBot: AI-Powered Ecological Assistant with Agentic Workflow
EcoBot is an intelligent ecological assistant that combines **multi-agent architecture** with **domain-specific tools** to provide scientifically accurate responses. Leveraging GPT-4o Mini and a sophisticated agentic workflow, it offers robust species identification, ecological analysis, and research paper processing capabilities.
---
## 🚀 Key Features
- **Agentic Workflow Architecture** (Planner → Evaluator → Executor)
- **Multi-Modal Analysis**: Images, PDFs, and text queries
- **Scientific Validation**: Wikipedia integration for taxonomic verification
- **Context-Aware Processing**: Maintains conversation context across sessions
- **BioTrove-CLIP Integration**: Specialized biological image classification
---
## 🧠 Architecture: Agentic Workflow
EcoBot implements a three-stage agentic pipeline to ensure accurate and reliable responses:
```mermaid
graph TD
A[User Input] --> B(Planning Agent)
B --> C{Requires Wikipedia?}
C -->|Yes| D[Wikipedia Tools]
C -->|No| E{File Type?}
E -->|Image| F[GPT4o-mini]
F --> H[Evaluating Agent]
E -->|PDF| G[PDF Processor]
G --> H[Evaluating Agent]
H --> B(Planning Agent)
E -->|Text| H[Evaluating Agent]
H --> I[Executing Agent]
I --> J[Response Generation]
```
### Core Components
1. **Planning Agent** (`planner.py`)
- Analyzes user intent using GPT-4o Mini
- Selects appropriate tools (Wikipedia/Image/PDF/GPT)
- Generates initial execution plan
2. **Evaluating Agent** (`evaluator.py`)
- Validates tool selection against domain rules
- Checks for recent duplicate queries
- Approves or revises execution plans
3. **Executing Agent** (`executor.py`)
- Orchestrates tool-specific operations:
- `wiki_tool.py`: Wikipedia API integration
- `image_tools.py`: BioTrove-CLIP + GPT-4o vision
- `pdf_tools.py`: Research paper analysis
- Maintains conversation context
- Formats final response with sources
---
## 📂 Directory Structure
```
namikazi25-ecobot/
├── app.py # Streamlit frontend
├── backend/
│ ├── agents/ # Core decision-making components
│ │ ├── evaluator.py # Plan validation
│ │ ├── executor.py # Tool execution
│ │ └── planner.py # Initial strategy
│ ├── tools/ # Domain-specific capabilities
│ │ ├── image_tools.py # Vision processing
│ │ ├── pdf_tools.py # Document analysis
│ │ └── wiki_tool.py # Fact verification
│ └── utils/ # Shared functionality
└── tests/ # Comprehensive test suite
```
---
## 🛠️ Installation & Usage
### Prerequisites
- Python 3.9+
- OpenAI API key
### Quick Start
1. **Clone repository**
```bash
git clone https://github.com/namikazi25/ecobot.git
cd ecobot
```
2. **Install dependencies**
```bash
pip install -r requirements.txt
```
3. **Configure environment**
```bash
echo "OPENAI_API_KEY=your_key_here" > .env
```
4. **Launch system**
```bash
# Start backend
cd backend && uvicorn main:app --reload
# In new terminal
streamlit run app.py
```
## 📚 Knowledge Base
EcoBot prioritizes scientific accuracy through:
- **GPT4o-mini Model**: Biological image classification
- **Wikipedia Verification**: Taxonomic data validation
- **Research Paper Analysis**: PDF text extraction + GPT synthesis
---
🤝 How to Contribute
We welcome contributions from developers, ecologists, and AI enthusiasts! Here's how you can help:
- Report Issues: Found a bug? Open an issue
- Suggest Enhancements: Have an idea? Start a discussion
- Submit PRs:
- Improve vision processing in image_tools.py
- Enhance PDF analysis in pdf_tools.py
- Expand ecological knowledge base
- First-time contributors are especially welcome!
---
"Empowering ecological exploration through intelligent agentic workflows" 🌍