Slack LLM Bot

Minimal Slack AI chatbot built with FastAPI and LLMs, deployable on Google Cloud Run with automated CI/CD via GitHub Actions.

---

Overview

This project provides a clean, production-ready foundation for building Slack chatbots powered by large language models.

It demonstrates:

• Real-time Slack event handling
• LLM-based response generation
• Containerized deployment on Cloud Run
• Automated deployment using GitHub Actions

---

Architecture

Slack -> FastAPI -> LLM -> Slack response

---

Project Structure

.
├── app.py
├── Dockerfile
├── requirements.txt
└── src/
    ├── llm.py
    └── slack_handler.py

---

Quickstart (Local)

1. Clone repository

git clone https://github.com/your-username/slack-llm-bot.git
cd slack-llm-bot

2. Create environment file

SLACK_BOT_TOKEN=your-token
SLACK_SIGNING_SECRET=your-secret
OPENAI_API_KEY=your-key

3. Run locally

uvicorn app:api --reload --port 3000

---

Deployment (Cloud Run)

Build and deploy manually:

gcloud builds submit --tag gcr.io/YOUR_PROJECT/slack-llm-bot

gcloud run deploy slack-llm-bot   --image gcr.io/YOUR_PROJECT/slack-llm-bot   --region europe-west1   --platform managed   --allow-unauthenticated   --set-env-vars SLACK_BOT_TOKEN=xxx,SLACK_SIGNING_SECRET=xxx,OPENAI_API_KEY=xxx

---

CI/CD (GitHub Actions)

This project uses GitHub Actions to automatically deploy to Google Cloud Run on push to main, withg the deployment pipeline:

• Trigger: push to main
• Build Docker image
• Push to Google Container Registry
• Deploy to Cloud Run
• Inject environment variables securely via GitHub Secrets

Required GitHub secrets:

GCP_PROJECT_ID
GCP_REGION
GCP_SA_KEY
SERVICE_NAME
SLACK_BOT_TOKEN
SLACK_SIGNING_SECRET
OPENAI_API_KEY

---

Usage

Invite the bot to a channel:

/invite @Slack LLM Bot

Send a message:

@Slack LLM Bot hello

---

Notes

• Slack retries events if responses exceed ~3 seconds
• Duplicate event handling should be implemented for production
• Cloud Run cold starts may introduce latency

---

License

MIT