## Weather Bot Agent Example with FunctionTool

examples/weather_bot/.env_sample

@@ -0,0 +1,4 @@
+OPENWEATHER_API_KEY=replace with yours
+GEMINI_API_KEY=replace with yours
+OPENAI_API_KEY=replace with yours
+LLM_MODEL=gemini/gemini-2.0-flash

examples/weather_bot/README.md

@@ -0,0 +1,124 @@
+# Weather Bot
+
+An intelligent weather assistant powered by AI that provides real-time weather information through both CLI and web interfaces.
+        ![bot](screenshorts/bot.png)
+## Features
+
+- 🤖 AI-powered conversational weather queries
+- 🌍 Real-time weather data from OpenWeather API
+- 💬 Dual interface: Command-line and Web UI
+- 🔧 Support for multiple LLM providers (Gemini, OpenAI)
+- ⚡ Fast and responsive
+
+
+## Installation
+
+### 1. Clone the Repository
+
+```bash
+git clone https://github.com/jentic/standard-agent.git
+cd standard_agent
+```
+
+### 2. Install Dependencies
+
+```bash
+make install
+```
+
+### 3. Activate Virtual Environment
+
+```bash
+source .venv/bin/activate
+```
+
+### 4. Navigate to Weather Bot
+
+```bash
+cd examples/weather_bot
+```
+
+### 5. Install Weather Bot Requirements
+
+```bash
+pip3 install -r requirements.txt
+```
+
+## Configuration
+
+### 1. Create Environment File
+
+Create a `.env` file in the `examples/weather_bot` directory:
+
+```bash
+touch .env
+```
+
+### 2. Add API Keys
+
+Add the following environment variables to your `.env` file:
+
+```env
+OPENWEATHER_API_KEY=your_openweather_api_key_here
+GEMINI_API_KEY=your_gemini_api_key_here
+OPENAI_API_KEY=your_openai_api_key_here
+LLM_MODEL=gemini/gemini-2.0-flash
+```
+
+### 3. Get API Keys
+
+- **OpenWeather API Key**: [https://home.openweathermap.org/api_keys](https://home.openweathermap.org/api_keys)
+- **Gemini API Key**: [https://aistudio.google.com/api-keys](https://aistudio.google.com/api-keys)
+- **OpenAI API Key** (optional): [https://platform.openai.com/api-keys](https://platform.openai.com/api-keys)
+
+## Usage
+
+### Command Line Interface (CLI)
+
+Run the weather bot in your terminal:
+
+```bash
+        python -m app.cli_bot
+```
+
+You can then interact with the bot by typing natural language queries like:
+- "What's the weather in London?"
+- "Will it rain in Tokyo tomorrow?"
+
+### Web Interface
+
+Start the web server:
+
+```bash
+        fastapi run app/bot.py 
+```
+
+The web interface will be available at:
+- **Local**: http://localhost:8000/chat
+- **API Docs**: http://localhost:8000/docs
+
+Open your browser and navigate to the local URL to interact with the weather bot through a user-friendly web interface.
+
+## Supported LLM Models
+
+You can configure different LLM models by changing the `LLM_MODEL` variable in your `.env` file:
+
+- `gemini/gemini-2.0-flash`
+- `gemini/gemini-1.5-pro`
+- `openai/gpt-4`
+- `openai/gpt-3.5-turbo`
+
+## Project Structure
+
+```
+examples/weather_bot/
+├── app/
+│   ├── agent.py          # CLI agent implementation
+│   └── ...
+├── requirements.txt       # Python dependencies
+├── .env                  # Environment variables (create this)
+└── README.md            # This file
+```
+
+
+![alt text](screenshorts/image.png)

examples/weather_bot/app/agent.py

@@ -0,0 +1,27 @@
+from dotenv import load_dotenv
+from agents.standard_agent import StandardAgent
+from agents.llm.litellm import LiteLLM
+from agents.memory.dict_memory import DictMemory
+from agents.reasoner.react import ReACTReasoner
+from .weather_tools import func_tools
+
+
+load_dotenv()
+
+
+# try changing to you own prefered model and add the API key in the .env file/ environment variable
+llm = LiteLLM(max_tokens=50)
+
+tools = func_tools
+memory = DictMemory()
+
+# Step 2: Pick a reasoner profile (single-file implementation)
+custom_reasoner = ReACTReasoner(llm=llm, tools=tools, memory=memory, max_turns=5)
+
+# Step 3: Wire everything together in the StandardAgent
+weather_agent = StandardAgent(
+    llm=llm,
+    tools=tools,
+    memory=memory,
+    reasoner=custom_reasoner,
+)