https://github.com/yash-aanand/taxbot
Taxbot is an AI chatbot that uses OpenAI Functions and Vercel AI SDK to help answer your tax-related queries.
https://github.com/yash-aanand/taxbot
chatbot next typre vercel-ai-sdk
Last synced: 3 months ago
JSON representation
Taxbot is an AI chatbot that uses OpenAI Functions and Vercel AI SDK to help answer your tax-related queries.
- Host: GitHub
- URL: https://github.com/yash-aanand/taxbot
- Owner: Yash-Aanand
- Created: 2025-03-26T11:45:12.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2025-03-26T16:08:43.000Z (over 1 year ago)
- Last Synced: 2025-03-26T17:24:51.458Z (over 1 year ago)
- Topics: chatbot, next, typre, vercel-ai-sdk
- Language: TypeScript
- Homepage:
- Size: 7.19 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Taxbot
Chat with Taxbot. Built with OpenAI Functions and Vercel AI SDK.
Introduction ·
Setting Up Locally ·
Tech Stack ·
Notes ·
## Introduction
This is an AI chatbot that uses OpenAI Functions and Vercel AI SDK to help answer your tax-related queries.
https://github.com/user-attachments/assets/79e9969f-60ef-455b-838b-2888e8ed1c35
## Setting Up Locally
To set up Taxbot locally, follow these steps:
1. Clone the repository
2. Install dependencies:
```bash
npm install
```
3. Create a `.env.local` file with your OpenAI API key:
```
OPENAI_API_KEY=your_api_key_here
```
4. Run the development server:
```bash
npm run dev
```
- open (http://localhost:3000/)
## Tech Stack
ChatH is built on the following stack:
- [Next.js](https://nextjs.org/) – framework
- [TanStack](https://tanstack.com/) - library
- [OpenAI Functions](https://platform.openai.com/docs/guides/gpt/function-calling) - AI completions
- [Vercel AI SDK](https://sdk.vercel.ai/docs) – AI streaming library
- [Vercel](https://vercel.com) – deployments
- [TailwindCSS](https://tailwindcss.com/) – styles
## Notes
Goals Met:
- Working Application fulfilling requirements
- Tech Stack Maintained
- Task 1 Completed
- Task 2 Completed
Assumptions Made:
- Regular GPT 3.5-turbo compatible openai api keys used
```
Text responses are preferred over multimodal responses, which can be adjusted accordingly
```
- No upper limit for prompts/hour (although spam prompting has been disabled)
```
A buffer already exists, but manually coding a breakpoint could completely avoid any complications
```
Placeholders/Simulations:
- File processing (in case no premium api keys available)
```
This can be expanded upon by fully utilizing the Multi-Modal AI SDK tools, alongside an OpenAI API key compatible with GPT-4o
```
- Suggestion Prompts
```
Could be expanded by hidden querying the bot to suggest prompts based on the message history for more useful prompts
```
## Author
- Yash Aanand - [LinkedIn](https://www.linkedin.com/in/yash-aanand-35192b273/) | [Website](https://yashaanand.com/)