Sheldon Bot

A Slack bot that listens to messages posted by the Orderli Bot in the #bot-urgent channel, sends the message text to a PostgreSQL MCP server, and replies in the same thread with the model's response.

Features

• Automatically responds to Orderli Bot messages in the configured channel
• Uses a PostgreSQL MCP server for generating responses
• Simple manual response tool for responding to specific messages

Setup

Environment Configuration

1. Create a dev.env file with the following variables:

   SLACK_BOT_TOKEN=xoxb-your-token
   SLACK_APP_LEVEL_TOKEN=xapp-your-token
   ORDERLI_BOT_ID=B06XXXXXXXX
   BOT_URGENT_CHANNEL_ID=C01XXXXXXXX
   MCP_SERVER_URL=http://postgres-mcp-server:8000/sse
   

   The scripts are configured to load environment variables from dev.env, not .env.

2. Make sure the MCP network is running (postgres-mcp-server)

Running with Docker

Build and run with Docker Compose:

docker-compose up --build

Running Locally

To run the manual response tool locally:

# Make sure you have the required dependencies installed
pip install -r requirements.txt

# Run the script
python respond_to_orderli.py

Usage

Automatic Responses

The bot will automatically listen for messages from Orderli Bot in the configured channel and respond to them.

Manual Responses

You can also manually respond to specific messages using the respond_to_orderli.py script:

python respond_to_orderli.py

This will:

1. Fetch recent messages from Orderli Bot
2. Let you select a message to respond to
3. Generate a response using the MCP server
4. Ask for your approval before posting

Docker Configuration

The docker-compose.yml file includes:

• Automatic restart configuration
• Environment variables loaded from dev.env
• Volume mapping for easy development

Requirements

• Docker and Docker Compose
• Slack app with appropriate permissions
• Access to a PostgreSQL MCP server