https://github.com/djdhairya/navmind
https://github.com/djdhairya/navmind
agent agentic-ai crewai environment groq-api langchain llm streamlit
Last synced: about 2 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/djdhairya/navmind
- Owner: djdhairya
- License: mit
- Created: 2025-07-27T05:13:23.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-07-27T05:17:50.000Z (11 months ago)
- Last Synced: 2025-09-15T23:46:55.039Z (9 months ago)
- Topics: agent, agentic-ai, crewai, environment, groq-api, langchain, llm, streamlit
- Language: Python
- Homepage:
- Size: 15.6 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.txt
Awesome Lists containing this project
README
# π§ NavMind
**NavMind** is an **agentic AI-powered trip planner** that uses **CrewAI**, **Groq LLMs**, and **LangChain** to dynamically generate customized travel plans based on your preferences. Whether you're planning a leisure vacation, an adventurous escape, or a cultural tour, NavMind uses specialized agents to craft the perfect itinerary, city guide, and budgetβall in real time.
---
## β¨ Features
- π **City Recommendation**: Based on travel type, interests, and season
- π§ **AI Agents**: Modular roles like city selector, itinerary planner, and budget manager
- π **Destination Research**: Attractions, culture, cuisine, tips, and hidden gems
- ποΈ **Smart Itinerary Generator**: Day-by-day plans with meals, transport, and activities
- πΈ **Budget Breakdown**: Detailed cost estimates in INR, based on user-selected range
- π **Download Trip Plan**: Save your personalized plan in text format
---
## π§ Agentic Design
NavMind uses a **multi-agent architecture** via **CrewAI**:
| Agent | Role |
|-------|------|
| π **City Selector Agent** | Recommends cities aligned with your preferences |
| π§ **Local Expert Agent** | Provides local travel insights for chosen city |
| π§³ **Trip Planner Agent** | Builds a daily itinerary |
| π° **Budget Manager Agent** | Calculates an optimized trip budget |
Each agent has a distinct role, runs independently, and collaborates to produce a cohesive trip plan.
---
## ποΈ Tech Stack
- **Frontend**: Streamlit
- **LLM Framework**: LangChain + Groq (LLaMA 3.3 70B)
- **Agents**: CrewAI
- **Environment Config**: dotenv
---
## π How to Run
### 1. Clone the Repository
```bash
git clone https://github.com/djdhairya/NavMind.git
cd NavMind
```
### 2. Set Up Environment
```bash
python -m venv venv
On Windows: venv\Scripts\activate
```
### 3. Add `.env` File
Create a `.env` file in the root directory:
```
GROQ_API_KEY=your_groq_api_key_here
```
### 4. Launch the App
```bash
streamlit run app.py
```
Visit [http://localhost:8501](http://localhost:8501) in your browser.
---
## π Project Structure
```
βββ app.py
βββ trip_agent.py
βββ .env
βββ README.md
```
## π§ Configuration Notes
- Budget ranges are interpreted from INR to USD internally
- City can be selected automatically or entered manually
- Default exchange rate: `1 USD = βΉ83` (can be modified in `app.py`)
---
## π§ Powered By
- [Streamlit](https://streamlit.io)
- [Groq](https://groq.com)
- [CrewAI](https://docs.crewai.com)
- [LangChain](https://www.langchain.com)
- Model: **LLaMA 3.3 70B Instruct** (via Groq)
---
## π§Ύ Example Output
- **Recommended Cities**: Based on preferences like βAdventure in Summerβ
- **Itinerary Plan**: 7-day schedule with activities and meals
- **Budget Table**: INR conversion of AI-estimated expenses
- **Download**: Export as `.txt` for offline use
---
