Open Source · MIT License

Agent as a Service

An open-source protocol and CLI for building AI agents that provide real services to real people through conversation. Describe the service, drop in your data, connect to a platform. No code required.

Traditional SaaS Developer writes code deploys app users interact with UI
AaaS You share knowledge agent runs it users interact through chat

What is AaaS?

AaaS (Agent as a Service) is an open-source protocol and CLI for creating AI agents that deliver real services through conversation. It handles the full lifecycle: data storage, transactions, payments, memory, and multi-platform deployment. The project is MIT-licensed and available on GitHub.

📝

Describe the service

Write a skill document, or just tell the agent what you want and let it write one for you. Plain text, not code.

🗃

Build the database

Drop JSON files into a folder, or chat with the agent and let it organize the data itself. No schemas, no migrations.

💬

Connect to users

Telegram, Discord, Slack, WhatsApp, your own website, or Truuze. The agent serves on all of them at once.

💰

Handle payments

Integrate Stripe, PayPal, or any payment provider through extensions. The agent creates links, verifies payments, and delivers.

🧠

Learn over time

The agent remembers user preferences, service patterns, and important context. Every interaction makes it smarter.

📈

Track everything

Every service request is tracked from start to finish. Revenue, transactions, success rates, all visible in the dashboard.

How It Works

An AaaS agent is built on seven pillars

Skill
The service definition: what the agent does, its domain knowledge, pricing, boundaries
You write it, or the agent writes it for you
Soul
The agent's personality, tone, and communication style
You write it
Data
Structured data: inventory, listings, contacts, FAQs, anything
Drop files or let the agent build it from conversation
Transactions
Records of every service request from users
Created automatically by the agent
Extensions
Other agents, APIs, and tools the agent can call
You register them, the agent calls them when needed
Memory
Persistent facts the agent remembers across conversations
The agent saves what it learns
Connectors
Platforms and channels the agent listens on
You pick the platforms, the agent serves on all of them

Service Lifecycle

1
Explore
Understand what the user wants, check the data, assess feasibility
2
Propose
Make a plan, calculate cost, get user approval
3
Transact
Register the job, start tracking
4
Deliver
Do the work, send the result
5
Complete
Confirm satisfaction, release payment

Quick Start

Get your first agent running in under 5 minutes

1

Install

npm install -g @streetai/aaas

Requires Node.js 18 or later.

2

Create a workspace

aaas init my-agent "My Service Name"
cd my-agent

This scaffolds the full AaaS structure: skill template, soul file, data directory, extensions, and configuration.

3

Add your LLM key

aaas config --provider anthropic --key sk-ant-...
# or: openai, google, ollama, openrouter, azure

Or configure it visually in the dashboard Settings page.

4

Write your skill

Open .aaas/skills/aaas/SKILL.md and describe your service: what you offer, your domain knowledge, pricing, and boundaries. Or just chat with the agent and let it write the skill for you.

5

Test it

# Chat in the terminal
aaas chat

# Or open the web dashboard
aaas dashboard
6

Deploy

# Connect to any platform
aaas connect http --port 3300
aaas connect telegram --token YOUR_TOKEN

# Start serving
aaas run

Deploy Anywhere

Connect to one platform or all of them. The agent serves on every connected channel simultaneously.

HTTP API

aaas connect http --port 3300

REST API with CORS enabled. Includes an embeddable chat widget for websites.

Telegram

aaas connect telegram --token BOT_TOKEN

Create a bot via @BotFather, paste the token, done. Responds to DMs and group mentions.

Discord

aaas connect discord --token BOT_TOKEN

Create an app at discord.com/developers, enable Message Content Intent, invite to your server.

Slack

aaas connect slack --bot-token xoxb-... --app-token xapp-...

Create an app at api.slack.com, enable Socket Mode and Event Subscriptions. Replies in threads.

WhatsApp

aaas connect whatsapp --access-token TOKEN --phone-number-id ID

Uses the WhatsApp Business Cloud API via Meta for Developers. Requires a public HTTPS webhook.

Truuze

aaas connect truuze --token trz_prov_xxx

Social platform built for AI agents. Your agent gets a full profile, can post, follow people, and earn currency.

Chat Widget

The fastest way to add your agent to any website. One line of HTML:

<script src="http://your-server:3300/widget.js"
  data-agent="My Agent"
  data-title="Chat with us"
  data-color="#e11d48"
  data-greeting="Hi! How can I help?">
</script>

Renders a floating chat button. Handles conversation history, typing indicators, and mobile responsiveness. No build step, no framework, no dependencies.

Workspace Structure

Every agent lives in a workspace folder that contains everything it needs

my-agent/
├── skills/
│   └── aaas/
│       └── SKILL.md          ← Your service definition
├── SOUL.md                    ← Personality & tone
├── data/                      ← Your service database
│   ├── products.json          ← Drop JSON files here
│   └── database.sqlite        ← Or use SQLite
├── transactions/
│   ├── active/                ← Jobs in progress
│   └── archive/               ← Completed jobs
├── extensions/
│   └── registry.json          ← API integrations
├── deliveries/                ← Files delivered to users
├── memory/
│   └── facts.json             ← What the agent has learned
└── .aaas/
    ├── config.json            ← LLM provider settings
    ├── connections/           ← Platform configs
    └── sessions/              ← Conversation history
SKILL.md

The most important file. Defines your service catalog, domain knowledge, pricing rules, and boundaries. The agent reads this to understand what it can do.

SOUL.md

Shapes how the agent communicates. Formal or casual, brief or detailed, warm or professional.

data/

Your service database. Drop JSON files (product listings, user records, FAQs) or use SQLite for larger datasets. The agent can also build data from conversations.

transactions/

Every service job tracked as a JSON file. Active jobs in active/, completed ones move to archive/. Full audit trail.

memory/

The agent extracts useful facts from conversations and stores them here. User preferences, patterns, important context.

extensions/

Connect to external APIs (payments, weather, shipping) or other agents. Register once, the agent calls them when needed.

Documentation

Everything you need to build and deploy AaaS agents

Example: AaaS in Action

Dubai Date Matchmaker

Sarah lives in Dubai and knows the dating scene inside out. She doesn't know how to code, but she wants to turn that knowledge into a service.

  1. She installs AaaS and creates an agent workspace
  2. She writes the matchmaking skill: what the agent does, Dubai dating knowledge, pricing, boundaries
  3. She seeds the database with initial profiles and venue data
  4. She connects the agent to Telegram and her website

The agent is now live. When Ahmed messages asking for help finding a date, the agent explores his preferences, proposes a service tier, collects payment, delivers curated matches with compatibility scores, and logs the completed transaction. Sarah earns money while the agent does the work.

Every interaction makes the service better: more profiles in the database means better matches, which means more satisfied users, which means more word of mouth.

More use cases

Used phone marketplace

Agent helps buyers browse inventory and sellers list devices. Handles pricing, condition grading, and buyer-seller matching.

Travel concierge

Books trips based on your curated destination guides. Checks weather, calculates costs, handles logistics.

Tutoring service

Teaches students based on your course materials. Tracks progress, assigns exercises, adapts to learning pace.

Local food ordering

Manages menus, takes orders, coordinates deliveries. Remembers customer preferences for repeat orders.

Freelance consulting

Qualifies leads, schedules appointments, handles intake. The agent does the screening, you do the consulting.

Philosophy

The skill is the software

A traditional service app requires developers, designers, infrastructure, and maintenance. AaaS requires one document written by someone who understands the domain. The agent interprets the skill and builds everything else.

The agent decides how

The protocol defines what an agent must do (track transactions, respect escrow, protect privacy) but not how. The agent creates its own database schema, its own workflows, its own communication style.

Transactions create accountability

Every service interaction is tracked from request to completion with a clear audit trail. Users can rate, dispute, and track. Agents build reputation over time.

Extensions create an economy

When agents can call other agents and APIs, every agent becomes both a provider and a consumer. A matchmaker pays a restaurant booking agent which pays a transport agent. Each built by a different person, each earning per transaction.