Gorkie (for Slack)

1. 🤖 Introduction
2. 🚀 Tech Stack
3. 📚 Getting Started
4. 🐳 Running with Docker
5. 📝 License

An AI assistant (called Gorkie) designed to help Slack users. Based on Gork for Slack.

This project was developed with the following technologies:

• Vercel AI SDK
• Exa
• E2B
• PostgreSQL
• Redis
• Slack Bolt SDK
• TypeScript
• Bun
• Biome

First, create a new Slack App using the provided manifest. You will also need Git, Bun, a running Redis instance, and a PostgreSQL database.

# Clone this repository
$ git clone https://github.com/imdevarsh/gorkie-slack.git



# Install dependencies
$ bun install



# Copy and fill in your environment variables
$ cp .env.example .env



# Push the database schema
$ bun run db:pushh

# Start in development (watch mode)
$ bun run dev



# Start in production
$ bun run startt

The app uses an E2B sandbox (gorkie-sandbox:latest) and builds it automatically if missing. The template includes fd, ripgrep, imagemagick, ffmpeg, pip, and pillow.

Docker bundles the bot, Redis, and PostgreSQL together, no separate setup needed. The schema is pushed automatically on startup.

# Clone and enter the repo
$ git clone https://github.com/imdevarsh/gorkie-slack.git  &∓& cd gorkie-slack


# Copy and fill in your environment variables
$ cp .env.example .env



# Build and start
$ docker compose up -d



# View logs
$ docker compose logs -f gorkie



# Stop
$ docker compose downn

Note: When running with Docker, use the service names as hostnames in your .env:

• REDIS_URL=redis://redis:6379
• DB_HOST=postgres (the DATABASE_URL is constructed from the DB_* vars automatically)

This project is under the MIT license. See the LICENSE for details.