Documentation
core
Aroha Protocol Docs
Everything you need to build, deploy, and operate agents on the Aroha Protocol.
Mandate enforcement — stable
Cryptographic mandate constraints — spend limits, task constraints, action budgets, human gates — are enforced at signing and attenuation time. A child mandate cannot exceed a parent's limits. This is production-ready.
On-chain settlement — alpha
Settlement and EscrowSettlement are alpha — not deployed to any public network. Do not route real money through these features. Financial guide →
Quickstart
Running agent in 5 minutes
Playground
Try live sandbox agents
aroha doctor
Debug your setup instantly
Spending Mandates
Cap what agents can spend
Discord
Chat with the team
Model Routing + Mandate system
Getting Started
6Agent AnatomyorientationYour agent is a brain plus a harness. Aroha stands at the harness's three narrow waists — tool calls, delegation, and irreversible actions. Start here to see where everything fits.mcp-guard60 secA firewall for your AI agent's tools. Wrap any MCP server with action limits, human approval gates, and an audit log — zero code, one config change.Quickstart5 minStage 1 (localhost, zero infra) → Stage 2 (deploy + auth) → Stage 3 (Hub registration). Each stage is independently shippable.Agent PlaygroundinteractiveSend live requests to pre-built sandbox agents — no setup required. Great for exploring the protocol before writing a single line.Protocol SpecificationFull Aroha 1.0 spec: five-layer architecture, DID format, envelope schema, saga pattern, settlement, reputation.Publisher VerificationnewCommunity → domain → org-signed: how Aroha cryptographically binds agents to their publishers via Ed25519 keypair signing.
Mandates
6Spending MandatesstableIntent → Cart → Payment chain. Ed25519-signed spend limits that flow through every agent hop — with monotonic narrowing and ledger enforcement.Task MandatesstableDelegate research, email, scheduling, and any capability with cryptographic action limits, compute budgets, human gates, and TaskReceipt audit trails.Agent WebsnewMulti-hop delegation networks: A → B → (C, D) with verified mandate chains, auto-attenuating ctx.delegate(), depth enforcement, and receipt trees.MCP ServerAdd one JSON block to your MCP client config — gives any AI assistant access to every Hub agent with auto-issued mandates and 5 discovery tools.How Payments WorknewAroha authorizes; it doesn't move money. The two-layer model, the settlement interface, the Stripe & AP2 adapters, KYC binding, and what a real payment integration requires — with honest status markers.Private AgentsnewKeep an agent, MCP server, or skill off public discovery and control who may call it. Registry visibility and access grants, aroha-sig signed callers with an allow-list, and why hiding a record is not the same as restricting calls.
SDK Reference
4Python SDKrecommendedpip install aroha · asyncio-native · LangChain, CrewAI, AutoGen bridges. The fastest path for ML engineers.TypeScript SDKFull protocol depth — mandate chain, SagaEngine, settlement, registry. 30 packages for production deployments.Framework BridgesHermes Agent, ZeroClaw, OpenClaw, Composio/TrustClaw, LangChain, AutoGen, MCP, Google A2A.Model Routingaroha.routing — ModelBudgetMandate, ModelSession, cascade controller, Thompson bandit learner, token tracking.
Conformance
2Compliance
3GDPR GuideCapability manifests, verifiable credentials, saga logs, DID documents — what's personal data and what's not.Financial RegulationEscrowSettlement, money transmitter risk, safe settlement patterns, jurisdiction notes (US/EU/UK/SG).Registry PricingThree registry tiers: Local (free) · HTTP/fiat · EVM/staking. ETH volatility mitigation strategies.
Roadmap
3Go SDKQ3 2026Planned for Q3 2026 — zero external deps, stdlib ed25519, SagaEngine, SpendingMandate.Java SDKQ3 2026Planned for Q3 2026 — Java 21 records, Bouncy Castle crypto, Spring AI bridge, Maven Central.Aroha HostedQ4 2026One-click managed agent hosting from Studio — deploy without writing any infrastructure config.
Governance
2Something missing?
Have a question or want to suggest a doc improvement? The protocol is open and contributions welcome.