Slack Kudos Bot

Made for Devfolio

A lightweight Slack bot for turning a message shortcut into a public kudos post.

What It Does

• Adds a message shortcut called Give Kudos
• Opens a modal so the sender can choose the recipient
• Either summarizes the thread with OpenAI or uses a custom message
• Posts a compact celebration message in a shared kudos channel

This repo does not currently persist kudos history or expose a slash command flow.

Privacy Notes

• Thread text is never written to application logs
• Slack mentions, IDs, emails, and links are sanitized before being sent to OpenAI
• If summarization fails, the bot falls back to a generic safe summary instead of echoing Slack text

Slack App Setup

1. Create the Slack app

1. Go to api.slack.com/apps
2. Create a new app from scratch
3. Install it into your workspace

2. Add bot scopes

Add these bot token scopes under OAuth & Permissions:

Scope	Purpose
chat:write	Post the kudos message
channels:history	Read public thread context
groups:history	Read private channel thread context
users:read	Resolve bot identity for self-protection checks

3. Enable Interactivity

Under Interactivity & Shortcuts:

1. Turn interactivity on
2. Set the request URL to https://your-domain/slack/events

4. Create the message shortcut

Under Interactivity & Shortcuts, create a Message Shortcut:

• Name: Give Kudos
• Short description: Celebrate someone for this thread
• Callback ID: give_kudos

5. Optional local Socket Mode

For local development, you can enable Socket Mode and create an app-level token with the connections:write scope. Set that token as SLACK_APP_TOKEN.

Environment Variables

Copy .env.example to .env for local development.

Variable	Required	Purpose
SLACK_BOT_TOKEN	Yes	Slack bot token
SLACK_SIGNING_SECRET	Yes	Verifies Slack requests in HTTP mode
KUDOS_CHANNEL_ID	Yes	Channel that receives kudos posts
OPENAI_API_KEY	Optional	Enables AI-generated thread summaries
SLACK_APP_TOKEN	Optional	Enables local Socket Mode

Running Locally

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python app.py

• If SLACK_APP_TOKEN is set, the bot runs in Socket Mode
• Otherwise it runs a local Flask server on http://localhost:3000

Deploying

Vercel

This repo exports a Flask app from app.py, which makes it compatible with Vercel's Python runtime.

1. Import the repo into Vercel
2. Set the environment variables above
3. Deploy
4. Point Slack interactivity to https://your-project.vercel.app/slack/events

Railway or other always-on hosts

The bot also works on Railway or similar Python hosts. In HTTP mode, the Procfile starts the Flask app with:

web: python app.py

Architecture

app.py         Flask + Slack Bolt app, shortcut handlers, local entrypoint
summarizer.py  Sanitized OpenAI thread summarization

License

MIT