jonbot

a friendly machine whats nice to talk to

This is a template repo, clone it and make your own friendo! Or use mine or whatever! No kings, no masters!

this bot helped me build this bot!

🤖❤️✨

---

jonbot is a chatbot with multiple frontends (Discord, Telegram) powered by AI through LangChain and connected to a MongDB database.

Installation

Create and activate an environment: e.g.

conda create -n jonbot-env python=3.11
conda activate jonbot-env

Clone the repo:

git clone https://github.com/jonmatthis/jonbot
cd jonbot

Install dependencies:

pip install -e .

Setup

Building yr Bot Buddy

Environment Variables

Setting up your project .env file

• Copy the sample.env file and rename it .env
• Add all the necessary API keys tokens, etc
  • OpenAI - https://beta.openai.com/docs/developer-quickstart/overview
    • make an OpenAI developer account - https://platform.openai.com/
    • then go to https://platform.openai.com/account/api-keys
    • Create new API key - it will look like sk-<a bunch of letters and numbers >
    • Copy the key and paste it into the OPENAI_API_KEY field in your .env file
  • OPTIONAL/RECOMMENDED - LangSmith (Helps you see what's going into and out of your bot's ai brain)
    • Sign up - https://smith.langchain.com (might be on waitlist?)
    • Make a new project and paste the project ID into the LANGSMITH_PROJECT field in your `.env file
    • Make a new API key and copy it into the LANGSMITH_API_KEY field in your .env file

Set up the bot's config .toml file

1. Choose a nickname for your bot

   • this can be anything and is only used behind the scenes (e.g. it doesn't affect the bot's displayed Username or anything).
   • I recommend something short, all lower case, and ending with the word 'bot', e.g. jonbot)

2. Copy this nickname into the BOT_NICK_NAMES filed of the .env file (if using multiple bots, separate them with a comma, e.g. BOT_NICK_NAMES=jonbot,otherbot,yetanotherbot)

3. Make a copy of the /jonbot/system/bot_tomls/_sample_config.toml file and rename it to [your_bots_nickname]_config.toml

   • You will fill in the DISCORD_TOKEN field after setting up the Discord Application (below)

Create a Discord Bot/Application

1. Go to https://discord.com/developers/applications
2. Click New Application
3. Give your application a name (this will be the name of your bot)
4. Click Bot in the left sidebar
   • Enable Privileged Gateway Intents (this is required for the bot to work)
     • Check all three tabs:
       • "Presence Intent"
       • "Server Members Intent"
       • "Message Content Intent"
   • Click Copy under DISCORD_TOKEN and paste it into the token field in your config file
5. Click OAuth2 in the left sidebar
   • Under OAuth2 URL Generator
   • in 'SCOPES' check bot
   • in 'BOT PERMISSIONS' check Administrator if you're lazy/reckless like me, or check the individual permissions you want your bot to have if you're nervous/careful about permissions
   • Copy the URL in the SCOPES box and paste it into your browser and invite your bot to your server

Software

• Install (and run) MongoDB Compass -https://www.mongodb.com/products/tools/compass

• Install (and run) Docker Desktop - https://www.docker.com/products/docker-desktop

• Install ffmpeg - https://ffmpeg.org/download.html

  • or brew install ffmpeg if you're on a mac
  • or sudo apt install ffmpeg if you're on linux
  • or choco install ffmpeg if you're on windows (requires chocolatey - https://chocolatey.org/install)

Run the Bot:

• Create/activate a python 3.11 environment
• install dependencies: pip install -e .
• run: python -m jonbot.__main__.py 🤖❤️✨
• The bot should now be running and you should see it online in your Discord server!

Run with Docker:

• cd into jonbot/docker
• run: docker-compose up --build -d
• The bot should now be running and you should see it online in your Discord server!

Usage

• Interact with the bot through Discord (text and voice memos)
• API endpoints provide programmatic access to core functionality
• __main__.py starts up all services

Architecture

graph TD
    classDef actor fill: #9E9E9E, stroke: #757575, stroke-width: 2px, color: #000, rounding: 5px;
    classDef interface fill: #90A4AE, stroke: #757575, stroke-width: 2px, color: #000, rounding: 5px;
    classDef api fill: #A5D6A7, stroke: #757575, stroke-width: 2px, color: #000, rounding: 5px;
    classDef core fill: #FFE082, stroke: #757575, stroke-width: 2px, color: #000, rounding: 5px;
    classDef data fill: #FFAB91, stroke: #757575, stroke-width: 2px, color: #000, rounding: 5px;
    A(["External User"]):::actor

    subgraph JonBot
        subgraph Frontend User Interfaces
            B1["Current:\nDiscord"]:::interface
            B2["Planned:\nSlack, WebApp, Android, etc"]:::interface
        end

        subgraph Backend
            C["API Interface (FastAPI)"]:::api

            subgraph Controller
                D["Controller"]:::core
                E["BackendDatabaseOperations"]:::core
                F["Core Processes:\n(Chatbot, etc)"]:::core
            end

            subgraph Data Layer
                G["Database MongoDB"]:::data
                H(["Pydantic Data Models"]):::data
            end
        end

    end
    A <--> B1
    A <--> B2
    B1 <--> C
    B2 <--> C
    C <--> D
    D <--> E
    D <--> F
    E <--> G

Loading

Layer 0 - Frontends

• discord_bot: Discord bot client and event handlers
• telegram_bot: Telegram bot client and handlers

Layer 1 - API Interface

• app.py: FastAPI application with endpoints

Layer 2 - Core Processes

• ai_chatbot: LangChain setup and processing
• audio_transcription: Transcribing audio to text with Whisper

Layer 3 - Data Layer

• database: MongoDB integration
• data_models: Pydantic models for data
• system: Configuration and utilities