Smart Agent Neo

Smart Agent Neo is a personal AI assistant platform with Morpheus decentralized inference as a first-class provider. It runs on your own devices and connects to the channels you already use: WhatsApp, Telegram, Slack, Discord, Google Chat, Signal, iMessage, Microsoft Teams, Matrix, and WebChat.

The core differentiator: decentralized AI inference via the Morpheus network, alongside centralized providers like Anthropic, OpenAI, Venice, and Ollama. Run open-source models (Kimi K2.5, GLM 4.7, Qwen3 235B, Llama 3.3 70B) through a P2P inference network with no single point of failure.

Morpheus Integration

Morpheus is supported in two modes:

Mode	How it works	Setup
Gateway (recommended)	API gateway at api.mor.org, OpenAI-compatible	Set MORPHEUS_API_KEY
Local proxy-router (advanced)	Fully decentralized, on-chain session management via MOR staking	Set MORPHEUS_ROUTER_URL + wallet

Available models on the Morpheus network:

Model	Context	Reasoning
Kimi K2.5	131K	No
Kimi K2.5 Web	131K	No
Kimi K2 Thinking	131K	Yes
GLM 4.7 Flash	202K	No
GLM 4.7	202K	Yes
Qwen3 235B	131K	Yes
Llama 3.3 70B	131K	No
GPT OSS 120B	131K	No

Models are dynamically discovered from the blockchain when using a local proxy-router. The static catalog serves as a fallback.

Full setup guide: docs/providers/morpheus.md

Install

Runtime: Node >= 22

npm install -g smart-agent-neo@latest

smart-agent-neo onboard --install-daemon

Works with npm, pnpm, or bun. The onboarding wizard walks you through gateway setup, workspace configuration, channel pairing, and provider auth.

From Source

git clone https://github.com/betterbrand/smart-agent-neo.git
cd smart-agent-neo

pnpm install
pnpm ui:build
pnpm build

pnpm smart-agent-neo onboard --install-daemon

# Dev loop (auto-reload on TS changes)
pnpm gateway:watch

Quick Start

# Start the gateway
smart-agent-neo gateway --port 18789 --verbose

# Talk to the assistant
smart-agent-neo agent --message "Ship checklist" --thinking high

# Send a message to a channel
smart-agent-neo message send --to +1234567890 --message "Hello from Smart Agent Neo"

Providers

Smart Agent Neo supports multiple inference providers. Configure via environment variables or the smart-agent-neo onboard wizard.

Provider	Auth	Notes
Morpheus	MORPHEUS_API_KEY	Decentralized inference via MOR network
Anthropic	ANTHROPIC_API_KEY or OAuth	Claude models
OpenAI	OPENAI_API_KEY or OAuth	GPT models
Venice	VENICE_API_KEY	Privacy-focused inference
Ollama	Local	Self-hosted models
Google	GEMINI_API_KEY	Gemini models
Amazon Bedrock	AWS SDK chain	AWS-hosted models
xAI	XAI_API_KEY	Grok models
Groq	GROQ_API_KEY	Fast inference
Together	TOGETHER_API_KEY	Open-source models
OpenRouter	OPENROUTER_API_KEY	Model aggregator

Providers are auto-detected from environment variables and auth profiles. Multiple providers can be active simultaneously with automatic failover.

Architecture

WhatsApp / Telegram / Slack / Discord / Google Chat / Signal / iMessage / Teams / Matrix / WebChat
               |
               v
+-------------------------------+
|           Gateway             |
|       (control plane)         |
|     ws://127.0.0.1:18789      |
+---------------+---------------+
                |
                +-- Pi agent (RPC)
                +-- CLI (smart-agent-neo ...)
                +-- WebChat UI
                +-- macOS app
                +-- iOS / Android nodes
                |
                v
+-------------------------------+
|       Inference Providers     |
|  Morpheus | Anthropic | OpenAI|
|  Venice | Ollama | Bedrock    |
+-------------------------------+

The Gateway is a local-first WebSocket control plane that manages sessions, channels, tools, and events. Inference requests are routed to whichever provider is configured, with automatic failover when multiple providers are available.

Key Features

• Multi-channel inbox -- WhatsApp, Telegram, Slack, Discord, Google Chat, Signal, iMessage (via BlueBubbles), Microsoft Teams, Matrix, WebChat
• Multi-agent routing -- Route channels/accounts to isolated agents with separate workspaces and sessions
• Voice Wake + Talk Mode -- Always-on speech for macOS/iOS/Android with ElevenLabs
• Live Canvas -- Agent-driven visual workspace with A2UI
• Browser control -- Managed Chrome/Chromium with CDP control
• Skills platform -- Bundled, managed, and workspace skills with install gating
• Companion apps -- macOS menu bar app, iOS node, Android node
• Sandbox mode -- Docker-based sandboxing for non-main sessions

Configuration

Minimal ~/.smart-agent-neo/smart-agent-neo.json:

{
  agent: {
    model: "morpheus/kimi-k2.5",
  },
}

Or use a centralized provider:

{
  agent: {
    model: "anthropic/claude-opus-4-6",
  },
}

Security

• Default: Tools run on the host for the main session (full access when it's just you).
• Group/channel safety: Set agents.defaults.sandbox.mode: "non-main" to run non-main sessions inside Docker sandboxes.
• DM pairing: Unknown senders receive a pairing code; approve with smart-agent-neo pairing approve <channel> <code>.

Run smart-agent-neo doctor to surface risky or misconfigured policies.

Chat Commands

Send these in any connected channel:

Command	Description
/status	Session status (model + tokens + cost)
/new or /reset	Reset the session
/compact	Compact session context
/think <level>	Set thinking level (off/minimal/low/medium/high/xhigh)
/verbose on|off	Toggle verbose output
/usage off|tokens|full	Per-response usage footer
/restart	Restart the gateway
/activation mention|always	Group activation toggle

Remote Gateway

The gateway runs well on a small Linux instance. Clients (macOS app, CLI, WebChat) connect over Tailscale Serve/Funnel or SSH tunnels. Device nodes (macOS/iOS/Android) handle local actions like camera, screen recording, and notifications via node.invoke.

Development Channels

Channel	Description	npm tag
stable	Tagged releases (vYYYY.M.D)	latest
beta	Prereleases (vYYYY.M.D-beta.N)	beta
dev	Head of main	dev

Switch with: smart-agent-neo update --channel stable|beta|dev

License

MIT