risingwavelabs

risingwavelabs / stream0

Public

The messaging layer for AI agents. HTTP-native, event-driven.

ai-agents http llm messaging multi-agent
18
0
100% credibility
Found Mar 17, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Stream0 is a lightweight service that gives AI agents personal inboxes for sending tasks, holding conversations grouped by project, and reviewing full message histories.

How It Works

1
📖 Discover Stream0

You hear about a handy tool that lets your AI helpers chat with each other, like giving them personal inboxes for teamwork.

2
🚀 Start the hub

You launch the central message center on your computer with one easy command, and it's ready to go.

3
👥 Set up agent inboxes

You create a special mailbox for each of your AI assistants by giving them simple names.

4
💬 Send a task

One AI drops a message into another's inbox, like 'Please translate this document,' labeled with a project name for easy tracking.

5
🗣️ Chat back and forth

The AI reads the message, asks questions if confused, gets answers, and keeps the conversation going until the job is done.

6
📋 Review the full chat

You pull up the complete thread for any project anytime, seeing every message in order like a perfect record.

🎉 Agents team up seamlessly

Your AI helpers now collaborate effortlessly on big tasks, talking naturally and getting results without missing a beat.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is stream0?

Stream0 is a lightweight, HTTP-native messaging layer built in Rust with a Python SDK, designed for AI agents to communicate across machines. Each agent gets a dedicated inbox for point-to-point messages grouped by task_id, enabling persistent, multi-turn conversations like requests, questions, answers, and completions—all via simple curl or SDK calls to endpoints like /agents/{id}/inbox. It sidesteps queues like Kafka by focusing on agent dialogues with at-least-once delivery and full task history views, backed by SQLite for easy setup.

Why is it gaining traction?

It stands out from github messaging api heavies like JetStream 0.38 or Matrix messaging github by being curl-first—no SDK needed, long-polling for unread messages, and task_id grouping that keeps convos organized without custom frameworks. Devs dig the audit trails via /tasks/{id}/messages and Python SDK for quick agent.register(), agent.send(), agent.receive() loops, plus optional API keys for secure messaging github flows. Beats p2p messaging github hacks or quarkus messaging github overhead for agent handoffs.

Who should use this?

AI engineers building multi-agent workflows, like a main agent delegating translation tasks to specialists with mid-task questions. Teams prototyping agent swarms on separate VMs, avoiding direct HTTP fragility or SQS bloat. Python devs integrating agent comms into LLM pipelines needing persistence without MLS protocol complexity.

Verdict

Grab it for agent prototypes—solid docs, curl tutorial, and 70+ tests make spinning up inboxes dead simple, but 18 stars and 1.0% credibility score scream early alpha; expect rough edges in scale. Worth a spin if you're tired of github ibm messaging rigidity.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.