Built for everyone
From managing a household to launching an enterprise AI strategy.
Easy
If you can describe what you need, you can build an agent. No code required. Visual dashboard. Templates for every field.
Secure
8 layers of Sammā Suit protection on every agent, every action, every message. Budget enforcement, audit trails, kill switches.
Powerful
Multi-agent councils that deliberate in parallel. 15 specialists analyzing your question before Sutra synthesizes the answer.
Affordable
BYOK (Bring Your Own Key) or use credits. Iceland servers: 100% renewable energy, 72% lower infrastructure costs. Savings passed to you.
Extendable
OpenClaw Skills Library compatible. Build custom skills. Connect any API. Every agent is a Portable Mind Format JSON — fork it, modify it, share it.
Your Choice
BYOK or credits. Claude, OpenAI, Google, DeepSeek, or local models. Your keys, your data, your agents. We never lock you in.
15 specialists. Day one.
An ethics council grounded in the Noble Eightfold Path, domain experts for every business function, and Sutra — who confers with all of them.
Strategic analysis & evidence evaluation
Motivation clarity & values alignment
Message evaluation & honesty-kindness balance
Ethical impact analysis & consequence modeling
Value creation vs. extraction analysis
Priority management & burnout detection
Pattern surfacing & blind spot detection
Deep analysis & single-problem immersion
Then build your own
Portable Mind Format (PMF) templates for a dozen fields. Open source. Fork, customize, deploy.
Portable Mind Format
Every agent is a structured JSON file — identity, voice, values, knowledge, behavioral constraints. It runs on Claude today. It runs on GPT tomorrow. The persona rides the model. Your agents are never locked to a single provider.
Read The Portable Mind"persona_id": "sutra-wisdom-judge",
"name": "Wisdom Judge",
"framework": "noble_eightfold_path",
"aspect": "right_view",
"voice": { "tone": ["analytical", "direct"] },
"model": "provider_agnostic",
"skills": ["web-search", "document-reader"],
"security": "sammasuit_8_layer"
}
OpenClaw Skills Compatible
Every skill in the ClawHub library works on Sutra.team — browser control, email, calendar, file management, Slack, Discord, and hundreds more. Plus every skill goes through Sammā Suit's SANGHA scanning before it touches your agents.
✓ Scanned by SANGHA · Approved · Installed
See it in action
Templates, skills, channels, audit trails, and agents working autonomously — all from your phone.

Agent Templates

Technical Templates

Agent Builder

Expert Agents

Skill Browser

Skill Details

Agent Config

Heartbeat & Channels

Telegram & Slack

Agent Memory

Audit Trail

The Promoter

Budget & BYOK

Agent Output

Launch Commander

Web Search Skill

Strategy Output

Agent Chat
8 layers of protection
OpenClaw has 512 known vulnerabilities. Sutra.team has Sammā Suit. Native platform gets all 8 layers. OpenClaw plugin gets 6.
Per-agent and per-council token budgets. Never overspend.
Every agent action logged. Full decision chain transparency.
Ed25519 keypairs per agent. Verify who said what.
AST scanning + governed network egress. No rogue skills.
Council-level emergency halt. Shut down all agents at once.
Per-agent model routing. Claude, OpenAI, Google, local.
Agents sandboxed during deliberation. No cross-contamination.
Hierarchical ethical principles guiding every response.
Already using OpenClaw? Install the Sammā Suit plugin for 6-layer protection. sammasuit.com
Powered by Iceland
Our infrastructure runs on 100% renewable geothermal and hydroelectric energy in Iceland — outside US surveillance jurisdiction, with the strongest privacy laws in Europe, and natural cooling that cuts costs by 72% compared to traditional US data centers. Those savings go directly to you.
Ask once. Hear everything.
Council deliberation is a skill your agents can invoke — or you can run it yourself from the dashboard.
Open-sourcing builds trust and community, but assess what you're giving away vs. keeping proprietary.
Patent protection covers the framework. SDK is safe to open-source under Apache 2.0 with a CLA.
Low risk if core IP stays closed. Moderate competitive risk — mitigate with fast iteration.
Open-sourcing security tooling aligns with industry responsibility.
Consensus: Open-source the SDK. Legal confirms patent protection. Ethics and Growth align on trust-building.
Tension: Risk Assessor flags competitive forking. Mitigation: fast iteration + community ownership.
Recommendation: Proceed with Apache 2.0 + CLA. Ship within 30 days.
Simple pricing
BYOK or credits. Start free, scale when ready.
Start with the full team. Build up to 5 agents.
Unlimited agents. Full power. All channels.
Iceland-hosted. Full power. Global privacy.
Multiple AI agencies. White-label. SSO. SLA.
Your agency is waiting
15 specialists. 32+ skills. 8 security layers. Open source. Start in minutes.
Get Started