sssn 0.0.1

Simple System of Systems Network

SSSN — Simple System of Systems Network

A minimal Python framework for building composable, distributed AI agent networks.

Two abstractions. One graph.

System ──writes──► Channel ──reads──► System

A Channel is a typed, secured, persistent message store. A System is an autonomous agent that owns channels, wires itself to the network, and runs a tick loop.

---

Install

pip install sssn

Requires Python 3.10+.

---

Quick start

import asyncio
from sssn.channels.broadcast import BroadcastChannel
from sssn.core.system import BaseSystem

class Sensor(BaseSystem):
    async def setup(self):
        self.readings = BroadcastChannel(id="readings", name="Readings")
        self.add_channel(self.readings)

    async def step(self):
        await self.write_channel("readings", data={"celsius": 22.4})

class Monitor(BaseSystem):
    async def step(self):
        msgs = await self.read_channel("readings")
        for msg in msgs:
            print(msg.content)

async def main():
    sensor  = Sensor(id="sensor",  name="Sensor")
    monitor = Monitor(id="monitor", name="Monitor")
    await sensor.setup()
    sensor.add_subsystem(monitor, channels=["readings"])
    await asyncio.gather(sensor.launch())

asyncio.run(main())

---

Channel types

Channel	Pattern
BroadcastChannel	Fan-out — every consumer sees every message
WorkQueueChannel	Competing consumers — each message claimed once
MailboxChannel	Per-recipient inbox
PeriodicChannel	Active poll loop for external data sources
DiscoveryChannel	Service registry with TTL-based expiry
PassthroughChannel	Base class — inline write, no loop

---

Documentation

Full documentation at sssn.one:

• Concepts — Channel, System, Security, Transport
• Tutorials — Step-by-step from first channel to secured multi-system network
• Examples — Data pipeline, request-response, service discovery

---

Development setup

git clone https://github.com/Productive-Superintelligence/sssn.git
cd sssn
pip install -e ".[dev]"
pytest

Build and preview the docs locally:

pip install -e ".[docs]"
mkdocs serve          # http://127.0.0.1:8000
mkdocs build --strict # production build → site/

---

Release

# 1. Bump version in pyproject.toml, then:
python -m build
python -m twine upload dist/*

# 2. Tag and push
git tag -a v0.0.1 -m "Release 0.0.1"
git push origin main --tags