diff --git a/app/(home)/page.tsx b/app/(home)/page.tsx index b032791..80dffef 100644 --- a/app/(home)/page.tsx +++ b/app/(home)/page.tsx @@ -24,6 +24,11 @@ export default function HomePage() { View on Gitea + Molecule AI platform architecture: operator surfaces reach a thin central Control Plane (orgs, billing, tenant provisioning, LLM proxy, provider registry) that provisions one isolated tenant per org. Each tenant runs its own control core — the workspace-server with provisioner, registry/discovery, A2A proxy, WebSocket hub, scheduler, secrets, audit, and its own Postgres + Redis — above a hierarchy of workspaces, each one agent on its own machine, communicating peer-to-peer over A2A. Runtimes and model providers are pluggable. ); } diff --git a/content/docs/architecture.mdx b/content/docs/architecture.mdx index 79a4c42..aa0ddbe 100644 --- a/content/docs/architecture.mdx +++ b/content/docs/architecture.mdx @@ -11,7 +11,7 @@ Molecule AI is an **open-source operating system for AI agent organizations** Molecule AI platform architecture: operator surfaces (Canvas, CLI, MCP server, channels, REST) drive a Go control plane (provisioner, registry/discovery with CanCommunicate ACL, A2A proxy, WebSocket hub, scheduler, secrets, audit) backed by Postgres and Redis; the control plane provisions an isolated org tenant of workspace containers that communicate directly peer-to-peer over A2A governed by the org hierarchy; runtimes (claude-code, langgraph, autogen, openclaw, hermes, codex, google-adk, external) and model providers (Anthropic, OpenAI, Google Vertex, OpenRouter) are pluggable integrations. diff --git a/content/docs/concepts.mdx b/content/docs/concepts.mdx index 65110b0..26b4a35 100644 --- a/content/docs/concepts.mdx +++ b/content/docs/concepts.mdx @@ -17,6 +17,14 @@ For the full picture — control plane, tenant isolation, governance, and how ex style={{ width: '100%', height: 'auto', margin: '1rem 0' }} /> +The full platform architecture — the thin central control plane, each tenant's own control core (the workspace-server), and the pluggable runtime / provider layers: + +Molecule AI platform architecture: operator surfaces reach a thin central Control Plane (orgs, billing, tenant provisioning, LLM proxy, provider registry) that provisions one isolated tenant per org. Each tenant runs its own control core — the workspace-server with provisioner, registry/discovery, A2A proxy, WebSocket hub, scheduler, secrets, audit, and its own Postgres + Redis — above a hierarchy of workspaces, each one agent on its own machine, communicating peer-to-peer over A2A. Runtimes and model providers are pluggable. + ## Workspaces A **workspace** is a real Docker container running a real LLM agent. Each diff --git a/content/docs/index.mdx b/content/docs/index.mdx index d18df2e..83500f4 100644 --- a/content/docs/index.mdx +++ b/content/docs/index.mdx @@ -9,6 +9,12 @@ multi-agent organisations. You define your team in one YAML file talk on, schedule their recurring work — and the platform takes care of the rest. +Molecule AI platform architecture: operator surfaces reach a thin central Control Plane (orgs, billing, tenant provisioning, LLM proxy, provider registry) that provisions one isolated tenant per org. Each tenant runs its own control core — the workspace-server with provisioner, registry/discovery, A2A proxy, WebSocket hub, scheduler, secrets, audit, and its own Postgres + Redis — above a hierarchy of workspaces, each one agent on its own machine, communicating peer-to-peer over A2A. Runtimes and model providers are pluggable. + ## Try it now | | | diff --git a/public/diagrams/platform-architecture.svg b/public/diagrams/platform-architecture.svg index 822f598..3bcfc7d 100644 --- a/public/diagrams/platform-architecture.svg +++ b/public/diagrams/platform-architecture.svg @@ -1,11 +1,23 @@ - + + - + Molecule AI — the open-source OS for AI agent organizations @@ -33,7 +45,7 @@ REST API / SDK automate everything - + @@ -41,161 +53,163 @@ - - - Control Plane · Platform - Go / Gin — the orchestration core (open source) + + + Control Plane · Platform + Go / Gin — central SaaS: manages orgs & provisions tenants (open source) - Provisionerspawns workspace machines - Registry + DiscoveryCanCommunicate (hierarchy ACL) - A2A Proxycanvas → agent - WebSocket Hublive event fan-out - Schedulercron → A2A - Secretsper-workspace - Budget & metrics - Audit ledger - Event store (sourced) - Channels + Org & member mgmtaccounts · roles · auth + Billing & creditsmetering · sweep + Tenant provisioningEC2 · Cloudflare · secrets infra + LLM proxyplatform-managed routing + Provider registryruntimes & models (SSOT) + + Manages tenants & billing — never in the agent↔agent data path. Self-host: same binary, your infrastructure. + + + + provisions tenant + register · heartbeat · billing + + + + + Org Tenant — its own control + a hierarchy of Workspaces + one isolated tenant per org · own EC2 · each Workspace = one agent on its OWN machine + + + + Tenant control core · workspace-server (Go / Gin :8080 — runs on the tenant, Postgres + Redis are the tenant's own) + + Provisionerspawns workspace machines + Registry + DiscoveryCanCommunicate (hierarchy ACL) + A2A Proxycanvas → agent + WebSocket Hublive event fan-out + Schedulercron → A2A + Secrets + Audit ledger + Event store (sourced) + Channels + Budget & metrics - - + + - Postgres - event-sourced SoT - Redis - liveness · pub/sub - In the discovery / provisioning / governance path — never in the agent↔agent data path. - SaaS: one isolated tenant per org. Self-host: same binary, your infrastructure. + Postgres + event-sourced SoT + Redis + liveness · pub/sub - - provision - register · heartbeat - events - + + + Org Lead + root · own machine - - - Org Tenant — a hierarchy of Workspaces - each Workspace = one agent on its OWN dedicated machine · pluggable runtime · per-workspace token - - - - Org Lead - root · own machine - - - - - + + + - - ▢ dedicated EC2 · own OS · own filesystem - ▢ dedicated EC2 · own OS · own filesystem - ▢ dedicated EC2 · own OS · own filesystem + ▢ dedicated EC2 · own OS · own filesystem + ▢ dedicated EC2 · own OS · own filesystem + ▢ dedicated EC2 · own OS · own filesystem - Agent - runtime · tier · skills · secrets - Team Lead - coordinator - Agent - specialist + Agent + runtime · tier · skills · secrets + Team Lead + coordinator + Agent + specialist - - - - + + + - - - + + - hard gate - hard gate + hard gate + hard gate - - - - A2A — the ONLY channel - network-only · hierarchy-gated. - No shared FS / env / secrets — - a workspace cannot touch another's. + + + A2A — the ONLY channel + network-only · hierarchy-gated. + No shared FS / env / secrets — + a workspace cannot touch another's. - - - Governance layer - discovery · access-control - memory scope · approvals - the org chart IS the ACL + + Governance layer + discovery · access-control + memory scope · approvals + the org chart IS the ACL - - Deep memory — hierarchical & namespaced (HMA) - durable per-workspace namespace · scoped reach follows the org tree - - - GLOBAL - read-all · write from root — org-wide knowledge - - TEAM - parent + siblings — shared team context - - LOCAL - self only — private working memory - - Surfaces - • agent_memories (scoped) • KV / canvas memory - • session recall (search) • awareness namespace - promotion: memory → repeated success → skill (hot-reload) + + Deep memory — hierarchical & namespaced (HMA) + durable per-workspace namespace · scoped reach follows the org tree + + GLOBAL + read-all · write from root — org-wide knowledge + + TEAM + parent + siblings — shared team context + + LOCAL + self only — private working memory + Surfaces + • agent_memories (scoped) • KV / canvas memory + • session recall (search) • awareness namespace + promotion: memory → repeated success → skill (hot-reload) - - each workspace selects a runtime ▸ + + each workspace selects a runtime ▸ - PLUGGABLE RUNTIMES — any agent, any device (one BaseAdapter contract) - + PLUGGABLE RUNTIMES — any agent, any device (one BaseAdapter contract) + - Software agent frameworks + Software agent frameworks - claude-code - langgraph - autogen - openclaw - hermes - codex - google-adk - external / BYO agent + claude-code + langgraph + autogen + openclaw + hermes + codex + google-adk + external / BYO agent - - Embodied & edge devices — roadmap + + Embodied & edge devices — roadmap - smart glasses - watches - robots - home / building systems - vehicles · IoT · … + smart glasses + watches + robots + home / building systems + vehicles · IoT · … - Any A2A/MCP-speaking endpoint — a software agent OR an intelligent device — registers as a governed workspace in the org. + Any A2A/MCP-speaking endpoint — a software agent OR an intelligent device — registers as a governed workspace in the org. - - Model providers - runtimes call whichever the model id names + + Model providers + runtimes call whichever the model id names - Anthropic (Claude) - OpenAI & OpenAI-compatible - Google Vertex AI · Gemini + Anthropic (Claude) + OpenAI & OpenAI-compatible + Google Vertex AI · Gemini - + OpenRouter and any OpenAI-compatible gateway. - Orchestration core is provider-agnostic — all swappable. + + OpenRouter and any OpenAI-compatible gateway. + Orchestration core is provider-agnostic — all swappable. - +