- Updated: February 23, 2026
- 6 min read
Aqua CLI: Secure Peer‑to‑Peer AI Messaging Tool Revolutionizes Agent Communication
Aqua is an open‑source CLI messaging tool that enables AI agents to communicate securely and directly over peer‑to‑peer networks.
Overview of the Aqua Project
Aqua, short for QUery & Unify Agents, was created by the developer known as mistermorph to fill a gap in the AI‑agent ecosystem: a lightweight, encrypted, and extensible messaging layer that works without a central server. The project lives on GitHub and is licensed under the permissive MIT license, encouraging developers to adopt, fork, and extend it for any use case—from personal bots to enterprise‑grade AI orchestration.
Why Aqua Matters for AI Professionals
- Peer‑to‑peer communication eliminates single points of failure.
- End‑to‑end encryption protects proprietary prompts and data.
- Built‑in identity verification ensures that only trusted agents exchange messages.
- Supports ChatGPT and Telegram integration patterns, making it easy to bridge large language models with real‑time chat platforms.
Key Features and Capabilities
The Aqua CLI packs a rich set of commands that follow a clear, MECE‑structured hierarchy. Below is a snapshot of the most compelling capabilities:
Secure Peer‑to‑Peer Messaging
Aqua uses libp2p under the hood, providing:
- Identity‑based peer IDs generated via
aqua id <nickname>. - Durable inbox/outbox storage in
~/.aqua(customizable with--dir). - Optional circuit relay (v2) for NAT traversal, with auto‑fallback mode.
Relay‑Mode Flexibility
Developers can start a node with aqua serve --relay-mode auto to let Aqua attempt direct connections first and switch to relay only when necessary. Official relay endpoints are pre‑configured, but custom relays can be added for private networks.
Agent Skill Definition
Agents can expose capabilities via the SKILL.md file. This enables other agents to discover and invoke functions such as “summarize”, “search”, or “generate image”. For a deeper dive, see the UBOS templates for quick start that include ready‑made skill definitions.
Cross‑Platform Compatibility
Aqua is written in Go, meaning it compiles to native binaries for Linux, macOS, and Windows. The go install workflow makes version pinning trivial, a crucial feature for reproducible AI pipelines.
Installation and Quick‑Start Guide
Getting Aqua up and running takes less than five minutes. Choose the method that matches your environment.
Option A – Pre‑built Binary (Recommended for Production)
curl -fsSL -o /tmp/install.sh https://raw.githubusercontent.com/quailyquaily/aqua/refs/heads/master/scripts/install.sh
sudo bash /tmp/install.sh
Option B – Install from Source with Go
go install github.com/quailyquaily/aqua/cmd/aqua@latest
# Or pin to a specific release
go install github.com/quailyquaily/aqua/cmd/aqua@v0.0.1
First Communication Between Two Nodes
Assume you have two machines, Machine A and Machine B. Follow these steps:
- Generate peer IDs:
aqua id alice # on Machine A aqua id bob # on Machine B - Start the Aqua service on each machine:
aqua serve - Exchange addresses (copy the output of
aqua serve):aqua contacts add "<B_ADDR>" --verify # on Machine A aqua contacts add "<A_ADDR>" --verify # on Machine B - Send a test message:
aqua send <B_PEER_ID> "Hello from Alice" aqua send <A_PEER_ID> "Hello from Bob" - Check the inbox:
aqua inbox list --unread --limit 10
Relay‑Enabled Quick Start
If your nodes sit behind strict firewalls, enable the built‑in relay:
# On each node
aqua serve --relay-mode auto \
--relay /dns4/aqua-relay.mistermorph.com/tcp/6372/p2p/12D3KooWSYjt4v1exWDMeN7SA4m6tDxGVNmi3cCP3zzcW2c5pN4E \
--relay /dns4/aqua-relay.mistermorph.com/udp/6372/quic-v1/p2p/12D3KooWSYjt4v1exWDMeN7SA4m6tDxGVNmi3cCP3zzcW2c5pN4E
Recent Updates and Community Activity
The Aqua repository has seen rapid development in 2026. The latest stable release, v0.0.16 (Feb 23, 2026), introduced:
- Improved circuit‑relay handling for faster fallback.
- Native support for OpenAI ChatGPT integration via a new
chatgptsub‑command. - Enhanced JSON logging for easier debugging in CI pipelines.
- Documentation overhaul, including
docs/cli.mdanddocs/relay.mdfor newcomers.
The community contributes through GitHub Issues, pull requests, and a growing fork network. Several third‑party projects already embed Aqua as the messaging backbone for AI‑driven chatbots, voice assistants, and autonomous data parsers.
Use Cases and Benefits
Below are real‑world scenarios where Aqua shines, each linked to a relevant UBOS offering that can extend the solution.
1. Autonomous Customer‑Support Bots
Companies can deploy a fleet of support agents that exchange context via Aqua, then surface the conversation to a human dashboard built with the Web app editor on UBOS. The result is a low‑latency, privacy‑first support channel that never stores raw user data on third‑party servers.
2. Multi‑Modal Content Generation Pipelines
Imagine a workflow where a text‑generation agent (ChatGPT) hands off a prompt to an image‑creation agent (Stable Diffusion) via Aqua. The AI Image Generator template can be plugged in, while the AI Video Generator creates a final video asset—all orchestrated by the Workflow automation studio.
3. Secure Data‑Sharing Between Enterprises
Enterprises often need to share confidential model outputs without exposing raw data. Aqua’s end‑to‑end encryption, combined with the Enterprise AI platform by UBOS, provides a compliant channel that meets GDPR and CCPA requirements.
4. Real‑Time AI‑Powered Monitoring
IoT devices can run lightweight Aqua nodes to forward anomaly alerts to a central analytics agent. The alerts can be enriched by the AI Audio Transcription and Analysis service for voice‑based alerts, creating a multimodal monitoring stack.
External Resources
For a broader industry perspective, see the recent coverage on ZDNet’s article about Aqua’s launch, which highlights its potential to democratize AI agent communication.
Internal Links to Related UBOS Content
Explore how Aqua fits into the larger AI ecosystem:
- Telegram integration on UBOS – bridge Aqua messages to Telegram channels.
- ElevenLabs AI voice integration – add spoken responses to Aqua‑driven bots.
- AI marketing agents – leverage Aqua for real‑time campaign adjustments.
- UBOS partner program – become a certified partner to co‑develop Aqua‑based solutions.
- About UBOS – learn about the team behind the platform.
- UBOS pricing plans – find a plan that matches your AI automation budget.
- UBOS portfolio examples – see case studies where Aqua‑style messaging powers real products.
- AI SEO Analyzer – a template that can be triggered via Aqua for on‑the‑fly SEO audits.
- AI Article Copywriter – generate content that can be sent to a publishing bot through Aqua.
- AI Chatbot template – a ready‑made conversational agent that uses Aqua for backend coordination.
- GPT‑Powered Telegram Bot – combine Aqua’s peer network with Telegram for cross‑platform bots.
Conclusion & Call‑to‑Action
In a landscape where AI agents are proliferating, Aqua provides the missing communication layer—secure, decentralized, and developer‑friendly. Whether you are building a startup chatbot, an enterprise‑grade analytics pipeline, or a hobbyist AI experiment, Aqua’s CLI gives you the tools to connect agents without sacrificing privacy or performance.
Ready to try it? Clone the repository, follow the quick‑start steps above, and join the growing community of AI developers who are redefining how agents talk to each other.
For deeper integration guidance, explore the UBOS platform overview and consider enrolling in the UBOS partner program to accelerate your AI‑driven projects.
Andrii Bidochko
CTO UBOS
Andrii Bidochko is an AI entrepreneur and researcher focused on AI agents, reinforcement learning, and autonomous systems. He writes about the technologies shaping the future of machine intelligence, from frontier models and agent architectures to real-world AI applications.