Aster

AI-powered Slack assistant with an easy skill system for companies. Teams can add custom "skills" — specialized knowledge and instructions — that shape how Aster responds, making it a configurable AI teammate.

Features

• Custom Skills — Add, update, remove, and search domain-specific skills per workspace
• AI-Powered Responses — Uses OpenAI (GPT-4o) with skill-aware context
• Slack Native — Slash commands, modals, @mentions, and DMs via Bolt SDK
• Multi-Tenant — Skills are isolated per Slack workspace
• Persistent Storage — SQLite database for skill persistence

Quick Start

# Install dependencies
npm install

# Copy environment template
cp .env.example .env
# Edit .env with your Slack and OpenAI credentials

# Run in development mode
npm run dev

# Run tests
npm test

# Build for production
npm run build
npm start

Slack App Setup

1. Go to api.slack.com/apps and create a new app
2. Enable Socket Mode and generate an App-Level Token (xapp-...)
3. Add the /aster slash command
4. Subscribe to bot events: app_mention, message.im
5. Add bot scopes: chat:write, commands, app_mentions:read, im:history
6. Install the app to your workspace and copy the Bot Token (xoxb-...)

Commands

Command	Description
/aster ask <question>	Ask Aster a question (uses all configured skills)
/aster skill add	Opens a modal to add a new skill
/aster skill list	Lists all skills for the workspace
/aster skill info <name>	View details of a specific skill
/aster skill remove <name>	Remove a skill
/aster help	Show available commands

What is a Skill?

A skill is a set of instructions and knowledge that customizes how Aster responds. For example:

• Code Review Expert — Instructions for reviewing code, what to look for, team standards
• Onboarding Guide — Steps for new employee onboarding, links, contacts
• Deploy Process — Your team's specific deployment checklist

Skills are added via a Slack modal with fields for name, description, instructions, and optional examples.

Environment Variables

Variable	Required	Default	Description
SLACK_BOT_TOKEN	Yes	—	Bot User OAuth Token (xoxb-...)
SLACK_APP_TOKEN	Yes	—	App-Level Token (xapp-...)
SLACK_SIGNING_SECRET	Yes	—	Signing Secret from app settings
OPENAI_API_KEY	Yes	—	OpenAI API key
OPENAI_MODEL	No	gpt-4o	OpenAI model to use
DATABASE_PATH	No	./data/aster.db	SQLite database file path
LOG_LEVEL	No	info	Log level (debug/info/warn/error)

Project Structure

src/
├── app.ts                          # Entry point, Bolt app initialization
├── config.ts                       # Environment config loader
├── types/index.ts                  # TypeScript types and Zod schemas
├── services/
│   ├── store.ts                    # SQLite persistence layer
│   ├── skills.ts                   # Skill business logic
│   └── ai.ts                       # OpenAI integration
└── listeners/
    ├── commands/aster.ts           # /aster slash command handler
    ├── views/skill-modal.ts        # Modal UI and submission handler
    └── messages/index.ts           # @mention and DM handlers

tests/
├── services/                       # Unit tests for services
├── listeners/                      # Command flow tests
├── views/                          # Modal structure tests
├── integration/use-cases.test.ts   # Multi-scenario integration tests
├── e2e/demo.test.ts                # 17-step end-to-end workflow demo
└── config.test.ts                  # Configuration tests

Development

npm run dev          # Start with hot-reload (tsx watch)
npm test             # Run all tests
npm run test:watch   # Run tests in watch mode
npm run test:coverage # Run tests with coverage report
npm run lint         # ESLint check
npm run lint:fix     # ESLint auto-fix
npm run format       # Prettier format
npm run build        # TypeScript compile

License

MIT