A Slack bot that processes app mentions by running Warp CLI agent commands. The bot can handle both direct mentions and threaded conversations, providing full context to the Warp agent.

• 📱 Responds to Slack app mentions
• 🧵 Processes entire thread context for threaded conversations
• 🚀 Runs in Docker for consistent environment
• 📁 Persists repository data between container restarts
• 🔧 Integrates with Warp CLI agent

The easiest way to set up your Slack app is using the provided manifest file:

1. Go to https://api.slack.com/apps
2. Click "Create New App"
3. Select "From an app manifest"
4. Choose your workspace
5. Copy and paste the contents of slack_app_manifest.json:
6. Install the app to your workspace

Copy the example environment file and fill in your tokens:

cp .env.example .env

Edit .env with your credentials:

Follow the instructions in .env for where to get relevant variables.

See https://docs.warp.dev/developers/cli#generating-api-keys for generating API key.

Create your repository configuration from the template:

cp repos.yaml.template repos.yaml

Then edit repos.yaml to specify which repositories to monitor:

repositories:
  - url: "username/repository-name"
    branch: "main"
  - url: "org/private-repo"
    branch: "develop"

Repository Configuration Options:

• url: GitHub repository in format owner/repo-name
• branch: Git branch to checkout and keep up to date as the 'base' branch for that repo

Examples:

repositories:
  - url: "mycompany/secret-api"
    branch: "production"

  - url: "mycompany/frontend"
    branch: "main"

The Slackbot uses a customizable prompt to process Slack messages and generate appropriate responses. Here's how to set it up:

1. See an example prompt - We use a comprehensive prompt at Warp to power our Slackbot that handles code questions and formulates PRs based on Slack discussions.

2. Use Warp Drive Notebooks - The Slackbot is designed to work with prompts stored in Warp Drive Notebooks. This allows you to dynamically iterate on your prompt without needing to re-deploy the bot.

3. Configure the notebook ID - To get your notebook ID:

   • Type @ in your Warp input
   • Select your notebook from the dropdown
   • Copy the alphanumeric code that appears after <notebook:{id}>
   • Add this ID to your .env file as the notebook reference

This setup enables you to refine your bot's behavior by editing the notebook prompt in real-time through the Warp Drive interface.

# Build and start the container
docker-compose up --build

# Run in background
docker-compose up -d

# View logs
docker-compose logs -f

# Stop the bot
docker-compose down

1. Invite the bot to channels where you want to use it
2. Mention the bot with @Warp your message here
3. Use in threads - the bot will include full thread context

Example interactions:

@Warp analyze the recent changes in the main branch
@Warp help me fix the bug discussed in the thread.