BriefChief is an intelligent Telegram bot that analyzes your team's chat conversations and automatically identifies task agreements, deadlines, and action items. It seamlessly integrates with Jira to provide context-aware insights powered by AI.

• 🔍 Smart Chat Analysis - Uses GPT-4o to analyze conversation history and extract task agreements
• 📋 Jira Integration - OAuth 2.0 authentication with full Jira API access
• 🌐 Multi-language Support - Built-in i18n support (English and Russian)
• 🔐 Secure Authentication - Token encryption and secure credential management
• 🛠️ LangChain Agent - Intelligent agent with Jira tools for context-aware responses
• 📊 Task Tracking - Automatically suggests task updates and status changes
• 💬 Natural Language Processing - Understands context and filters out casual conversations

┌─────────────────┐      ┌──────────────────┐      ┌─────────────────┐
│  Telegram Bot   │─────▶│  LLM Handler     │─────▶│    OpenAI       │
│  (bot.py)       │      │  (call_llm)      │      │    GPT-4o       │
└────────┬────────┘      └────────┬─────────┘      └─────────────────┘
         │                        │
         │                        │ Tools
         │                        ▼
         │               ┌──────────────────┐
         │               │  Jira Tools      │
         │               │  (LangChain)     │
         │               └────────┬─────────┘
         │                        │
         ▼                        ▼
┌─────────────────┐      ┌──────────────────┐
│  Jira Auth      │◀────▶│  Jira Cloud API  │
│  Server         │      │  (OAuth 2.0)     │
└─────────────────┘      └──────────────────┘

• Python 3.10 or higher
• Telegram Bot Token (from @BotFather)
• Telegram API credentials (from my.telegram.org)
• OpenAI API Key
• Jira Cloud account with OAuth 2.0 app configured
• PostgreSQL or file-based storage for tokens

git clone https://github.com/vladmiron85/BriefChief.git
cd briefchief

pip install -r requirements.txt

For the Jira Auth Server:

pip install -r jira_auth_server/requirements.txt

Create a .env file in the root directory:

# Telegram Bot Configuration
TELEGRAM_BOT_TOKEN=your_bot_token_here
TELEGRAM_API_ID=your_api_id
TELEGRAM_API_HASH=your_api_hash
TELEGRAM_SESSION_STRING=your_session_string

# OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key

# Jira Auth Server URL
JIRA_AUTH_SERVER_URL=http://localhost:5000

# Internal API Key (for secure communication)
INTERNAL_API_KEY=your_generated_internal_api_key

Create .env file in jira_auth_server/ directory:

# Jira OAuth 2.0 Configuration
JIRA_CLIENT_ID=your_jira_client_id
JIRA_CLIENT_SECRET=your_jira_client_secret
JIRA_REDIRECT_URI=https://yourdomain.com/auth/callback
JIRA_BASE_URL=https://auth.atlassian.com

# Token Encryption
ENCRYPTION_KEY=your_fernet_encryption_key

# Internal API Key (must match bot's key)
INTERNAL_API_KEY=your_generated_internal_api_key

Generate Telegram Session String:

python generate_session.py

Generate Encryption Key:

python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())"

Generate Internal API Key:

python -c "import secrets; print(secrets.token_urlsafe(32))"

1. Go to Atlassian Developer Console
2. Create a new OAuth 2.0 (3LO) app
3. Set callback URL: http://localhost:5000/auth/callback (for local development)
4. Add permissions: read:jira-work, read:jira-user, write:jira-work (the last one is dangerous. don't switch it on during experiments until you 100% sure that your AI agent is fine)
5. Copy Client ID and Client Secret to your .env file

Terminal 1 - Start Jira Auth Server:

cd jira_auth_server
python jira_auth_server.py

Terminal 2 - Start Telegram Bot:

python bot.py

• /auth - Authenticate with Jira
• /status - Check your Jira authentication status
• /brief - Analyze recent chat messages and identify task agreements
• /test - Test with pre-generated chat history

1. Authenticate with Jira:

   User: /auth
   Bot: 👋 Please authorize with Jira...
   [User clicks button and completes OAuth flow]
   Bot: ✅ Authorization Successful!
   

2. Analyze Chat:

   User: /brief
   Bot: Select LLM model:
   [User selects OpenAI]
   Bot: Getting chat messages...
   Bot: Processing messages with OpenAI...
   
   Found task agreements:
   
   Task: Update login page design
   State change: New status is In Progress
   Comment to add: Team agreed to use blue color scheme...
   Task link: PROJ-123
   

briefchief/
├── bot.py                      # Main Telegram bot
├── messages.py                 # i18n message definitions
├── generate_session.py         # Telegram session generator
├── requirements.txt            # Python dependencies
├── .env.example               # Example environment variables
│
├── LLM/                       # LLM and AI logic
│   ├── __init__.py
│   ├── llm_handler.py        # LLM orchestration
│   ├── jira_tools.py         # Jira LangChain tools
│   ├── prompt_system.txt     # System prompt
│   └── prompt_user.txt       # User prompt template
│
└── jira_auth_server/         # Jira OAuth server
    ├── jira_auth_server.py   # Flask auth server
    ├── requirements.txt       # Auth server dependencies
    ├── .env.example          # Example auth server config
    ├── JIRA_AUTH_README.md   # Detailed auth docs
    └── auth_flow_diagram.md  # OAuth flow diagram

Edit LLM/prompt_system.txt and LLM/prompt_user.txt to customize how the AI analyzes conversations:

# prompt_system.txt
You are an AI assistant specialized in analyzing chat conversations...

# prompt_user.txt
Your task is to:
1. Read through the provided chat history
2. Identify any agreements made on tasks
3. Summarize changes to be made
...

Edit messages.py to add new language support:

MESSAGES = {
    "ru": { ... },
    "en": { ... },
    "es": {  # Add Spanish
        "auth_button": "🔐 Autorizar con Jira",
        ...
    }
}

• Check if both services are running (bot and auth server)
• Verify TELEGRAM_BOT_TOKEN is correct
• Check logs for errors

• Verify OAuth app configuration in Atlassian
• Check JIRA_CLIENT_ID and JIRA_CLIENT_SECRET
• Ensure callback URL matches exactly

• Verify OPENAI_API_KEY is valid and has credits
• Check OpenAI API status
• Review prompt templates for syntax errors

• Regenerate session string with generate_session.py
• Verify TELEGRAM_API_ID and TELEGRAM_API_HASH

This project is licensed under the MIT License - see the LICENSE file for details.

• python-telegram-bot - Telegram Bot API wrapper
• LangChain - LLM application framework
• OpenAI - GPT-4 API
• Telethon - Telegram client library