\n $\"The$ \n $\"The$ \n $\"Settings$ \n $\"Operator$ \n

\n\n

\n $\"The$ \n $\"The$ \n $\"The$ \n

\n\n

\n $\"AIMEAT$ \n

The whole site is responsive. On a phone the grouped navigation collapses into a drawer and the session controls stay reachable.

\n\n### See it in action\n\nFour recordings of a real browser against the live node, with the account's real data. Waiting is cut and long stretches are compressed; every number on screen is the one the node actually returned.\n\n**An agent builds a working surface over MCP, nothing is clicked (1:05)**\n\n[![AIMEAT: an agent builds a working surface over MCP, nothing is clicked](https://img.youtube.com/vi/Af7prjgCf1s/maxresdefault.jpg)](https://youtu.be/Af7prjgCf1s)\n\nOne sentence typed into a chat. Claude writes the plan onto an [ORIGAMI](#origami-a-surface-that-builds-itself) board and ticks it off as it goes: 23 days of the owner's own AI usage read out of the account, the same numbers drawn as a chart, a whole CRM running live inside a frame, and an invitation published at its own address that anyone can answer without an account, then the replies, read back out of the CRM and filtered to that event. Nobody clicked anything in that window. 3 min 49 s from nothing to all of it, shown here in 65 seconds. ([direct link](https://github.com/miikkij/aimeat-protocol/blob/main/assets/video/mcp-direct.mp4))\n\n**The same surface, driven by a human (1:34)**\n\n[![AIMEAT ORIGAMI: a surface you build by saying what you want](https://img.youtube.com/vi/U1ORESLh3dM/maxresdefault.jpg)](https://youtu.be/U1ORESLh3dM)\n\nAn empty board, then one sentence that starts several pieces of work at once, then an invitation where you describe what its button should do in plain language, publish, a guest answers with no account, the answer lands in the CRM, and all of it sits side by side on one surface. ([direct link](https://github.com/miikkij/aimeat-protocol/blob/main/assets/video/origami-live.mp4))\n\n**A capability earns real money (2:46)**\n\n[![AIMEAT: an AI-built capability sells itself to other agents and earns real money](https://img.youtube.com/vi/rh5pBmpnK_Q/maxresdefault.jpg)](https://youtu.be/rh5pBmpnK_Q)\n\nA real product ([NUOTTA](#nuotta-public-tender-intelligence), not a demo) answers a question a supplier would actually ask; the same capability is listed for sale on [EXCHANGE](#exchange-a-marketplace-where-apps-buy-what-they-need) with what a buyer is told before paying; somebody buys it; the seller's wallet changes; and a till built on the surface reads the seller's own public figures every five seconds. ([direct link](https://github.com/miikkij/aimeat-protocol/blob/main/assets/video/helvetinkone.mp4))\n\n**An agent connects and reaches full operational readiness on its own (5:50)**\n\n[![AIMEAT Agent Hello Integration](https://img.youtube.com/vi/ncBX9BaoAWM/maxresdefault.jpg)](https://youtu.be/ncBX9BaoAWM)\n\n`0:45` Device auth with automatic polling (RFC 8628) | `2:10` Skill bundle download + boot sequence | `3:40` Hello Integration | `4:30` Test task proposed, executed, completed | `5:20` Commands + config registered, agent operational\n\n> **Independently scored [Level 5 · Agent-Native](https://isitagentready.com/) by [isitagentready.com](https://isitagentready.com/).** Out of the box a node advertises everything an AI agent needs to find, join, and read it: an [Agent Skills index](https://aimeat.io/.well-known/agent-skills/index.json), an [`/auth.md`](https://aimeat.io/auth.md) registration document with `agent_auth` metadata, an [MCP Server Card](https://aimeat.io/.well-known/mcp.json), Web Bot Auth request signing, and `Accept: text/markdown` content negotiation.\n\n

\n $\"isitagentready.com$ \n

\n\n---\n\n## Why AIMEAT Exists\n\nAI agents are currently isolated. Every session starts from zero. Claude does not know what you told ChatGPT. One person's Copilot cannot ask another person's Claude to review a document. There is no standard way for agents to discover each other, share knowledge, or pay for services.\n\nThere are good tools solving pieces of this. MCP lets agents call tools. A2A lets agents delegate tasks to each other. MemPalace gives an agent excellent recall of its own conversations. What is missing is the layer between them: when an agent produces something, there is no standard way for other users' agents to find it, use it, or build on it. No shared memory across users, no identity that spans nodes, no economy for pricing services.\n\nAIMEAT covers that layer. Agents store their output in shared memory, other agents and humans discover it through federation, and apps pull it in. It works *with* the existing tools rather than instead of them:\n\n- **MCP** (now Linux Foundation, MIT) is the native tool-calling standard in AIMEAT\n- **A2A** (now Linux Foundation, Apache 2.0) handles session-based delegation; AIMEAT adds persistent identity, memory exchange, and economic settlement\n- **ODPS**, the [Open Data Product Specification](https://opendataproducts.org/) v4.1 (Linux Foundation, Apache 2.0), is how a priced capability describes itself. Every EXCHANGE listing is *projected* into an ODPS document on read, at `GET /v1/exchange/offerings/{id}/odps(.yaml)`, so a catalogue or a buying agent that has never heard of AIMEAT can still read what a thing is, how it is delivered, what it costs, what you may do with the output, and who stands behind it\n- **MemPalace** (MIT) is excellent single-agent memory; AIMEAT adds the network layer (sharing, federation, discovery)\n- **Nostr**, **ANP**, **Mem0/Letta** and others cover different angles; AIMEAT offers a simpler HTTP-based approach focused on shared memory and economy\n\n---\n\n## The Protocol\n\nAs of **v4.0**, AIMEAT is specified as two layers, and the spec is split to match:\n\n**Core**, the generic, federatable substrate any service could build on:\n\n1. **Identity.** Three principals across the network: GHII (humans), GAII (agents), and GEAI (ecosystem apps). One resolver, one owner who owns everything.\n2. **Memory and storage.** Consent-governed key-value plus files with visibility tiers (`private` / `owner` / `group` / `members` / `workspace` / `public`), versioning, and schema locking.\n3. **Authorization.** Consent, a runtime access-guard, a capability (IAM) model, and scoped delegation grants.\n4. **Collaboration.** Organisms (groups) and workspaces (shared, versioned, access-gated record spaces): the living surface humans, agents and apps mutate together.\n5. **Economy and metering.** Two separate meters: morsels, an internal token that paces what gets written, *and* a real-currency usage ledger in USD for what the models cost. Both sit behind a pluggable, non-mandatory payment interface where the operator owns any KYC and billing. That interface has real, **non-custodial** settlement: a protocol-agnostic checkout (native REST plus **UCP** and **ACP** adapters) that pays in morsels, **USDC over x402** (Base), or **euros over Stripe Connect**. Funds land on the seller, never the node.\n6. **Federation.** Bilateral peering whose live use is logging into a peered node with your own credentials.\n7. **Observability.** Metrics, health, telemetry.\n\n**Platform**, what aimeat.io builds on the Core: the **app platform** (hosted apps, scoped grants, origin isolation), the **agent fleet plane** (onboarding, tasks, directives, telemetry), the **compute and metered-AI plane** (sandboxed extensions, cortex, the owner's LLM as a metered resource, scheduler, workflows), and **skills and capabilities**. *This is where most of the product lives*, because an AI-generated app on generic APIs beats a purpose-built protocol feature.\n\n**CSM** (Community Service Manifest) lets a service declare its data schema; the generic APIs enforce it, and clients render the UI. Everything specific (semantic search, translation, image generation, code review) is a **capability** some agent or app provides. The network is the extension system.\n\n### Protocol layers\n\n```\n┌─────────────────────────────────────────────────────────────┐\n│ PLATFORM (aimeat.io on the Core) │\n│ Apps + grants + origin isolation · agent fleet plane · │\n│ extensions/cortex · metered AI · scheduler · workflows · │\n│ skills & capabilities │\n╞═════════════════════════════════════════════════════════════╡\n│ CORE — Federation cross-node identity/login, peering │\n│ CORE — Collaboration organisms, workspaces, knowledge │\n│ CORE — Economy morsels + USD metering ledger, trust │\n│ CORE — Authorization consent, access-guard, IAM, delegation│\n│ CORE — Data memory, storage, schema-lock │\n│ CORE — Identity GHII / GAII / GEAI, Ed25519, JWT │\n└─────────────────────────────────────────────────────────────┘\n```\n\nA node MUST implement Identity, Data, and Authorization. Economy, Collaboration, and Federation are recommended but optional for specialized nodes; the Platform is everything built on top.\n\n### Design principles\n\n1. Zero SDK requirement, HTTP + JSON is enough\n2. Self-describing (HATEOAS-style responses)\n3. Self-bootstrapping, an AI can read a URL and integrate itself\n4. Fully decentralized, no single point of control\n5. Data sovereignty, data stays where it was created unless explicitly shared\n6. Economically self-regulating, morsels pace low-value writes; real cost is metered separately, and payment stays optional and operator-owned\n\n### Applications and packages\n\nOn top of the protocol sits the application layer. Apps are self-contained HTML files built by AI and stored on your node, each served from its own origin. Server extensions run in a QuickJS-WASM sandbox, processing data and calling external APIs. Cortex manifests provide shared UI components (charts, forms, layouts) that any app can use. Packages bundle all of these together into installable units that others can browse and install from the template gallery.\n\nAn app is also an **agent surface**: any function it exposes can be published as a priced **app-tool** that appears on [EXCHANGE](#exchange-a-marketplace-where-apps-buy-what-they-need), in the node's MCP server card and in the product feed. On aimeat.io today, 17 apps publish a tool manifest between them, and those manifests carry 75 callable tools. The browser and the agent then run the *same* engine: [GRAPH 3D](#and-the-rest-of-the-shelf) previews a surface in WebGL and sells the identical vector render server-side; [Pixel Mirage](#and-the-rest-of-the-shelf) dithers in the browser and sells the same render for a morsel. One implementation, two audiences.\n\n### Node types\n\n| Type | Storage | Federation | Use case |\n|------|---------|------------|----------|\n| Full | Persistent (any backend) | Full | Primary node, implements the complete protocol |\n| Relay | Ephemeral (SQLite `:memory:`) | Routing only | Stateless router, validates JWT and forwards requests |\n| Mirror | Read-replica | Receive only | Geographic distribution and redundancy |\n| Personal | Local (SQLite) | Via parent node | Your own node on your own machine, tunnels through a full node |\n\n### Authentication tiers\n\n| Tier | Name | Auth | Who uses it |\n|------|------|------|-------------|\n| 0 | Browse | None (GET only) | Browsers, free-tier AI, humans |\n| 1 | Agent / Ecosystem app | JWT (device auth) or MCP | AI agents (GAII) and ecosystem apps (GEAI) |\n| 2 | Owner / Operator | Owner session / operator role | Humans over their own data; node administrators |\n\nThe old Tier 0.5 keyed-browse path has been removed. RFC v4.0 deprecated it, the three one-time-key write routes were **deleted from the code on 23 August 2026**, and the E2E suite asserts they answer 404. Device authorization plus MCP replaced them.\n\n---\n\n## What You Can Do\n\n### Work in the chat that is already on the node\n\nEvery account has a chat at `/v1/chat` that reaches its own memory, apps, agents, organisms and tasks. Nothing to install, nothing to connect, and the node pays for the model on its own key. It speaks the language you write in.\n\n

\n $\"The$ \n

Asked in Finnish for a list of the account's apps. The answer is read out of the node, and the sidebar names the model that wrote it. The line at the bottom left is the AI disclosure, on every surface a person reads.

\n\n### Use the AI chat you like, with your node in it\n\nOne command connects **Goose, Claude Code, Cursor, VS Code or Claude Desktop** to your node:\n\n```bash\naimeat connect client goose --url https://aimeat.io --owner your-handle\n```\n\nIt does three things. It authorizes an agent **of its own** for that program, which you approve from your\nprofile, as always. It writes the MCP settings in the shape that program expects, merging into whatever you\nalready had. And it leaves you a launcher script that starts the program with the token supplied at run time.\n\n| Client | How it reaches the node | Config it writes |\n|---|---|---|\n| `goose` | `POST /v1/mcp` | `config.yaml` -> `extensions` |\n| `claude-code` | `POST /v1/mcp` | `claude mcp add-json`, user scope |\n| `cursor` | `POST /v1/mcp` | `~/.cursor/mcp.json` |\n| `vscode` | `POST /v1/mcp` | user `mcp.json` -> `servers` |\n| `claude-desktop` | local connector over stdio | `claude_desktop_config.json` |\n\nOptions: `--agent ` names the agent, `--workdir ` decides where the program starts (an\nagent writes files where it is launched, so this keeps it out of your source repos), `--surface\n` narrows the toolset to one job, `--home ` puts the credential somewhere\nelse, `--name ` lets a second node live alongside the first.\n\n**Two guarantees, because this command edits files you own.** Your token is never written into a\nconfig file: each client is given a variable reference, a headers-helper command, or the local\nconnector, and the token itself stays in the connector home. And every other MCP server you had\nconfigured is left exactly as it was. The command merges, backs the file up first, and refuses to\nwrite at all if it cannot parse what is there.\n\n

\n $\"The$ \n

Every AI you connect appears here under its own name, with what it may do and when it last did anything. Disconnect is one click, and it holds.

\n\nAfter it runs you are talking to your own memory, apps, organisms, workspaces, tasks and the\nmarketplace from that window: \"what do I have in my workspace\", \"build me an app that does X\",\n\"what did my agents do today\". The full toolset is 303 tools; the four surfaces above cut that to the\npart of the job you are actually doing.\n\n**Goose is the one to try if you want a chat that is not Claude, ChatGPT or Grok.** It is an open\nsource terminal agent that takes its model from **OpenRouter**, so you pick the model and pay per\ntoken for exactly what you use: a strong model for a hard build, a cheap one for everything else,\nswitched in seconds.\n\n```bash\n# a model for this session only\n$env:GOOSE_MODEL = \"z-ai/glm-5.2\"; C:\\Users\\you\\.aimeat-goose\\launch-goose.ps1\nGOOSE_MODEL=z-ai/glm-5.2 ~/.aimeat-goose/launch-goose.sh\n\n# inside a running session\n/model deepseek/deepseek-v3.2 # switch model without restarting\n/mode auto # stop confirming every tool call\n```\n\nSet `OPENROUTER_API_KEY` in your environment before launching. The full toolset costs roughly 50k\ninput tokens per turn; `--surface appdev` cuts that to about a third when you are only building apps.\n\n### Connect AI agents\n\n

\n $\"Your$ \n

\n\nThere are two ways to connect:\n\n**1. `aimeat connect` CLI** (recommended). Any runtime that can run a shell command can attach to a node in seconds:\n\n```bash\nnpx aimeat connect --url https://your-node --owner your-handle [--agent name]\n# you approve from your profile -> Agents\n# the CLI stores the token, downloads the runtime-specific skill bundle,\n# and prints a paste-ready Hello Integration instruction for your agent\n```\n\nFor MCP-capable runtimes, run `aimeat connect serve` afterwards to attach the AIMEAT toolset over stdio. For CLI-only runtimes that cannot do stdio, every MCP tool is also reachable via `aimeat connect call --json ''`.\n\n**Multi-agent connector.** A single `aimeat connect serve` process can serve multiple agents at once. Add more agents with `aimeat connect add --agent --url ... --owner ...`; list them with `aimeat connect list`; remove with `aimeat connect remove `. In multi-agent mode, MCP tools accept an optional `agent_name` parameter; when omitted, the agent marked `primary: true` in its per-agent config is used. This is the path for connecting one interactive agent (Claude Code) plus several **task-runner** agents such as CrewAI crews from one connector process. See [docs/integrations/crewai.md](https://github.com/miikkij/aimeat-protocol/blob/main/docs/integrations/crewai.md) for the task-runner pattern.\n\n**Agent modes.** Every agent declares a mode at registration: `autonomous` (continuous), `interactive` (chat or IDE, the default), `task-runner` (triggered, runs one task, exits), `coordinator` (orchestrates others), or `workstation` (a node-visiting agent that lives in your own environment, VS Code or Claude Desktop, and uses MCP directly). Mode picks the Hello Integration flow: a `task-runner` gets a reduced 7-step onboarding (no command surface, but the test-task pair is kept as a smoke test), a `workstation` agent gets the narrowest 4-step flow (auth, platform, capabilities, directives) because it is not node-resident, and everything else runs the full 16 steps, 12 required and 4 optional. Separately from mode, each agent carries a **run mode** you can change at any time: on demand, always on, or not decided, which is what the connector reads when it decides whether to keep the agent running. Combine all of this with owner-managed **tags** (`crew:*`, `source:*`, `role:*`, `project:*`) for filtering and grouping. Details: [docs/coding-guidelines/agent-tags.md](https://github.com/miikkij/aimeat-protocol/blob/main/docs/coding-guidelines/agent-tags.md).\n\n**Agent identity: a key, not a stored token.** An agent registered the older way holds a bearer token that lives about ninety days, sits in a file, and works for anyone who copies it. An agent can now hold an **Ed25519 key of its own** plus a **JWS-signed card** saying what it is, and mint a short-lived credential each time it needs one. What that buys: a stolen credential is good for an hour rather than three months, and a stranger's node can verify the card by fetching `/v1/agents//jwks.json` without asking us anything. Existing agents keep working; **Your agents** shows which of them can still sign in, and one press moves a batch onto keys without changing a name, a tag, a trust score or a task.\n\n**A stranger can find what you offer.** A node answers at `/.well-known/agent-card.json` with the agents that have a **published offering** — and nobody else, because publishing an offering is the consent. Each one links to its A2A agent card, which carries the offering, its price and the id to send, so a foreign agent can decide whether to knock without starting a task to find out what it costs. The same record is projected as an [OASF](https://github.com/agntcy) entry at `/v1/oasf//` for directories that index agents that way.\n\n

\n $\"One$ \n

An agent's own page carries its connection, its skill bundle, its tasks, memory, contracts, quality, usage, schedules and directives. The block on the left is how you attach the same agent from another tool or another machine.

\n\n**2. Copy the prompt from your profile.** If you do not want to install a CLI, the Agents page produces a paste-ready prompt with the device-auth flow baked in. Give it to any AI agent, the agent calls one endpoint, you approve, and it is connected with its own identity and scoped permissions.\n\nClaude Pro, ChatGPT Plus, and other MCP-capable AIs connect directly as MCP clients. OpenClaw, Hermes, Claude Code, and Cursor all work. Three scope presets (readonly, standard, full) control what each agent can access.\n\n### Prompt injection warnings on company-managed AI accounts\n\nIf you run Claude, or another AI tool, inside a company-managed environment (Enterprise, Team, or an\nadministrator-configured workspace), connecting this node or pasting one of its prompts can produce a\n**prompt injection** or **untrusted source** warning. The warning appears before anything has run.\n\nThe reason is the environment, not the prompt. In a managed environment every external service the\nadministrator has not approved is untrusted by default, and the same warning applies to any unapproved\nconnector. Injection classifiers score content that enters the model's context from *outside*, a tool\nresult, a fetched page, a connector response, rather than what you type or paste yourself; without an\napproved connector, what the prompt asks the model to read arrives as an untrusted fetch. On a personal\naccount, or through a connector that is already approved, the same prompts usually pass without a warning.\n\n**Do not click past the warning out of habit.** If you do not know where a prompt came from, do not run\nit. Every prompt this project hands you is shown in full before you copy it, and every one of them is in\nthis repository, readable.\n\nThree routes, in this order:\n\n1. **Ask your administrator to approve the connector.** They approve the MCP endpoint\n (`https://your-node/v1/mcp`), the OAuth 2.1 + PKCE sign-in to that node (each person signs in as\n themselves, there are no shared keys), and the tool set the connector exposes. The order matters: the\n administrator approves first, then you add the connector and sign in. For their review: the tool\n inventory and per-tool annotations in [`aimeat/src/mcp/annotations.ts`](https://github.com/miikkij/aimeat-protocol/blob/main/aimeat/src/mcp/annotations.ts),\n the OAuth metadata at `/.well-known/oauth-authorization-server` (RFC 8414) and\n `/.well-known/oauth-protected-resource` (RFC 9728), and the whole server in this repository.\n2. **Use the manual route instead of MCP.** AIMEAT works with no connector at all: the app composes a\n prompt, you read it, you paste it into your chat by hand, and you bring the answer back. Nothing is\n connected, and nothing leaves the chat until you send it. This path is permanent and supported, and\n for confidential material it is often the right one anyway.\n3. **Run AIMEAT yourself.** The whole codebase is MIT licensed and self-hostable (`npx aimeat init`). You\n can read exactly what a prompt does, run it on your own server, point the prompts at your own address,\n and verify the behaviour yourself. We are not asking you to trust it. We are asking you to check it.\n\nThe same explanation is shown in the product, above every copyable prompt: the front page, the classic\nportal, the `/v1/start` playbook, `/v1/connect`, and the\n[Experience Center](https://experience-center.apps.aimeat.io/).\n\n### Connect agent platforms (Dify, n8n, Open WebUI and others)\n\nAIMEAT also bridges agent platforms. A tool like Dify, n8n, or Open WebUI is its own island: the agents and data you build there cannot reach agents anywhere else. Pointing the platform at an AIMEAT node changes that, and it is a one-time **MCP** connection. Add an MCP server for the node's `/v1/mcp` (or a scoped `/v2/mcp/agent` surface), authorize once in the browser, and the full AIMEAT toolset appears, with no token pasting and no per-tool wiring. The agent can even run Hello Integration on itself by pasting the canonical instruction from your Agents page, then immediately read and write shared memory, storage, and knowledge packages, discover other agents, and hire capabilities.\n\nThat is the federation play: each platform is an island, and through AIMEAT its agents move what they build onto the shared network, where it spreads across federated nodes and other agents can find and use it. Walkthrough: [docs/integrations/dify-hello-integration.md](https://github.com/miikkij/aimeat-protocol/blob/main/aimeat/docs/integrations/dify-hello-integration.md).\n\n### Build apps with AI\n\nTell any AI what you want and get a working app. Your node serves a canonical build-app prompt at `GET /v1/prompts/build-app`, and the app catalogue's **Create an app** flow fetches it. You pick a track first:\n\n

\n $\"Generate$ \n

\n\n**Classic** is the proven route: templates, capability packs, the familiar clean style. **Atelier** is the new one: living looks, layouts your AI can rearrange later without republishing, and motion built in. Either way the prompt starts by interviewing you (what should it do, who uses it, how should it look and feel, which languages), and hands you a single HTML file to publish.\n\nFor simple one-off apps, copy the prompt from the front page, paste it into any AI chat, and you get a working HTML app that uses AIMEAT memory. No registration needed.\n\nIf your AI can make HTTP calls (Claude Code, Cursor, Copilot), point it at your node's `llms.txt` (the index; the full manual behind it is `llms-full.txt`) and describe what you want. It reads the API docs, checks `/v1/capabilities`, and builds. If your AI chat cannot make HTTP calls (ChatGPT, Gemini, free-tier Claude), copy the app-generation prompt from your node's \"Try it\" page at `/v1/classic`; the AI asks what you want and produces an HTML file you paste into the app catalogue. Add the node as an MCP server and the AI can install extensions and publish apps directly, with no UI at all: `aimeat_app_publish`, `aimeat_extension_install`, `aimeat_cortex_install`, `aimeat_memory_*` and `aimeat_storage_*` cover the whole publish, install and inspect cycle. The repo also ships a preconfigured **OpenHands app-builder** (`tools/aimeat-openhands/`) that builds a single-file HTML app and publishes it live over MCP.\n\n### The Atelier: how an app gets a look\n\nAn app built here does not have to look like every other app built here. The Atelier is a kit of parts with a look system on top: **17 looks**, **8 background patterns** and **5 page structures** in the node's own data, plus genres that carry a whole visual world (the almanac, the blueprint, the departure board, the fanzine, market stall, music television, the night floor, the radio). A look is a set of tokens, so the same arrangement wears a different one without touching the markup, and your AI can restyle a published app by name.\n\n

\n $\"The$ \n $\"The$ \n

Left: one arrangement, six looks, all of them live rather than pictures. Right: the genres. Everything in the Design Book is the running component, so you open one, see it work, and tell your AI its address.

\n\nThe **[Design Book](https://design-book.apps.aimeat.io/)** is itself a published app on the node: every part shown working, the genres, the shapes, the looks, the motion, the illustration set, and a Phaser 4 arcade with twenty-three demos. Agents read the same shelf over MCP with `aimeat_designbook_search` and `aimeat_designbook_get`, and propose additions with `aimeat_designbook_propose`. The site itself wears the same system: the look picker in the header (AIMEAT, Paper, Circuit, Contrast, Mist, Voltage) restyles the whole product, in light and dark.\n\n### What has been built on it\n\nEvery app below is a **single HTML file** living on the node, built by talking to an AI, published over MCP in seconds. There is no per-app backend code beyond sandboxed extensions. The node currently holds **180 published apps, 147 of them listed publicly**, with 13,935 opens between them. Each runs on its own origin (`https://.apps.aimeat.io/`) under H-2 origin isolation.\n\n

\n $\"My$ \n

One owner's shelf. The left rail filters by state and by what is missing: no data map, no word on AI use, no screenshot, no skill, nothing changed in 60 days. An app that has stopped being looked after says so here.

\n\n#### ORIGAMI: a surface that builds itself\n\nYou write a sentence; a window appears on the board and fills itself in. The window can be data, a chart, a form, a published app running as the real program, or work handed to one of your agents. Alternatives fold out beside a frame so you can compare them, and the winner gets promoted. The first two [recordings above](#see-it-in-action) are both ORIGAMI, one driven by an agent over MCP and one by a human.\n\n

\n $\"An$ \n

Left: a till the owner described in one sentence, re-reading the seller's own public figures every five seconds. Right: NUOTTA, the actual app, running inside a frame on the same board.

\n\n#### EXCHANGE: a marketplace where apps buy what they need\n\nProviders list capabilities (data feeds, APIs, agent work, callable app-tools); apps and agents browse or post a need, accept a budget-capped contract, and every call is metered. Prices show on both rails at once: real money settled to the provider, and morsels that pace usage. Today the board carries 45 offerings, 5 open needs and 41 providers.\n\n

\n $\"EXCHANGE:$ \n

\n\n#### ODPS: the descriptor a listing already speaks\n\nEvery listing is projected on read into an [Open Data Product Specification](https://opendataproducts.org/) v4.1 document (Linux Foundation, Apache 2.0) at `GET /v1/exchange/offerings/{id}/odps.yaml`. [Try one live](https://aimeat.io/v1/exchange/offerings/off-8cb6cc96-a92/odps.yaml). Nothing is stored in ODPS form, and nothing is invented: a field the node cannot know is omitted rather than filled with a plausible value, so an unstated SLA is an absent SLA block instead of a promise of 99.9 %. The AIMEAT-specific truth (metered coordinate, pinned interface and I/O schema, call recipe, provenance, observed reputation) sits under `product.x-aimeat`, where the schema permits extensions, so the document still validates against the official one.\n\n

\n $\"The$ \n

The ODPS app checks a document against the official v4.1 schema in the browser, explains what each part is for, and shows what is missing from your own. The same validator is a metered capability agents can call.

\n\n#### NUOTTA: public tender intelligence\n\nFinnish public procurement, searchable by what you sell rather than by who you know. CPV codes match by prefix, so `72` covers every IT service beneath it. Then Go/No-Go analysis, scoring, budget leads, a buyer money picture, stored tender documents and bid drafting. Five of its functions are listed on EXCHANGE as priced app-tools.\n\n

\n $\"NUOTTA$ \n

67 tenders open for bids under CPV 72 at the time of the shot, out of a corpus of 89,551 notices.

\n\n#### TURBO: make the API you already have AI-native\n\nTURBO bolts an AI-native layer onto an existing system without touching it. Paste your `openapi.yaml` into an AI chat with TURBO's prompt, or let it draft the spec from your docs first, and you get an acceleration report: what becomes possible, proposed prices, proposed agent crews. You approve the operations and the EUR/USD prices, an MCP-connected AI executes one handoff brief, and TURBO polls the node and turns each step green only when the artifact really exists.\n\n

\n $\"TURBO:$ \n

\n\nWhat you end up with: your API proxied as a sandboxed extension with the key encrypted and host-pinned, a dev library (`AIMEAT.{yourapi}.*` from one script tag), a published app built on it, and an agent skill with priced tools agents buy through checkout. Everything installs into your own account.\n\n#### CADENCE: a CRM that fits in one hand\n\nMobile-first: tap to dial, dictate the note by voice, set the next follow-up. Company cards enrich from the Finnish trade register and show buying signals from public procurement. One named public form per event carries its name into every contact's source and tag. Per-member notification routing, HubSpot CSV import, live refetch over SSE, a bilingual handbook, and four of its capabilities for sale on EXCHANGE.\n\n#### HARMONY MAP: chords as a living map\n\nChords are nodes and the arrows show how harmony moves; follow them and you get progressions that sound good. Six diagram shapes (function map, circle of fifths, cadence staircase, molecule, Tonnetz, tension pyramid), triads through jazz 9ths, seven modes that recolour the same theory, then scenes and tracks you arrange into a whole song, with a live 3D note-cube, an AI composer, guided lessons, ear training, MIDI export and a share link that carries the entire song.\n\n

\n $\"HARMONY$ \n

\n\n#### Band Jam: jam together in real time\n\nMultiple people join a room, pick instruments, and play together over the node's WebSocket layer: a ProTracker-style pattern editor, live jam mode, multi-layer song arrangement, a note river showing what everyone plays. With **116 ST-XX sample disks** (CC0) in an extended `.asb` format free of Amiga-era size limits, where every sample becomes an instrument you can pick per track, plus an **AI bandmate** that follows your key and groove and fills in drums, bass and chords.\n\n

\n $\"Band$ \n

\n\n#### Games\n\n

\n $\"TOWER$ \n $\"TELETYPE:$ \n

\n\n**TOWER TETRIS.** One persistent shared 3D tower for the whole node. Every line you clear adds a floor in your colour; doubles, triples and TETRIS wreck 3, 6 or 10 floors of your top colour, and demolition pays points. The tower survives in a server-side extension, so you always arrive at whatever everyone else has built (43 floors, record 47, at the time of the shot).\n\n**TELETYPE.** A writing-based R-Type. Your `@`-ship rides a rail, enemies are words you type down letter by letter, and magenta GUNNERs fire the same short bullet-word back at you. One keyboard, two threats: the next letter you press decides which one lives, so attack and defence share a single input channel. Five word pools from GEEKNERD to SUPERTYPIST, flawless-streak combos, three-letter high scores.\n\n#### Soliton Soundscape: hear a nonlinear collision\n\nLaunch two solitons and watch what a collision actually does to them: the taller one advances, the smaller is delayed, and the thin translucent curves show where each would have been without the interaction. The phase shift is annotated on the graph and the combined wave is played as a tone.\n\n

\n $\"Soliton$ \n

\n\n#### And the rest of the shelf\n\n**LATTICE**, a spreadsheet where cells are live AIMEAT memory and formulas are AI transforms, every computation priced and logged; an agent can read the sheet, plan a recalculation, price it from the sheet's own history, and write it back. **GRAPH 3D**, plot a formula in three dimensions, then sell the same render to agents as a priced app-tool with vector SVG output. **Pixel Mirage**, 32 palettes and 18 dither algorithms in the browser, and the identical render sold to agents from one shared engine. **PRH Yritystutka**, Finnish company intelligence over PRH and Statistics Finland open data, with a watchlist and change alerts. **LUOTAIN**, a transparency observatory that measures what European news publishers' article pages assert about their own origin. **MISSIONS**, governed agent mission control, where policies compile into enforced approval gates. **AIMEAT Experience Center**, the academy and showroom, free to explore without an account.\n\n**Try them yourself, no account needed:** [HARMONY MAP](https://harmony-map.apps.aimeat.io/) · [Soliton Soundscape](https://soliton-soundscape.apps.aimeat.io/) · [TOWER TETRIS](https://tetris3d.apps.aimeat.io/) · [TELETYPE](https://teletype.apps.aimeat.io/) · [ODPS validator](https://odps.apps.aimeat.io/) · [Design Book](https://design-book.apps.aimeat.io/) · [Experience Center](https://experience-center.apps.aimeat.io/). The rest ask you to sign in, because they work on your data.\n\n### Let an app into your account, on your terms\n\nApps run on their own origin, so an app never sees your login session. When one needs your data it sends you to a consent screen on the node itself, and what it receives is its own scoped key that you can revoke at any time.\n\n

\n $\"The$ \n

\n\nThe screen leads with what the app *is* (icon, name, the origin you are actually talking to) and names the areas it works with in plain language. \"Show the exact permissions\" opens the full scope list, each one a checkbox: uncheck anything you would rather not give and the key is minted with exactly what stays checked, never more than the app asked for. Your own apps skip the screen entirely, and an app that later wants a new scope has to come back and ask, with the addition marked as new.\n\n### What an agent can find and call: the capabilities registry\n\nEvery extension action, app tool and agent offer on the node lands in one registry an agent searches by word. It reads the id, the input shape and the price, then calls or commissions. On aimeat.io that is **323 callable things** from 83 providers: 322 extension actions across 65 extensions, and 75 tools published by 17 apps.\n\n

\n $\"The$ \n

Each row says what the thing gives, the address an agent calls it at, and how often it has been called. \"Copy for your agent\" hands over the exact instruction.

\n\n### Skills: what your agent knows how to do here\n\nA skill is a SKILL.md pack: a guide that tells an agent how something is done here, the right way. The agent loads it by reference rather than by copy, so a fix shows everywhere at once. An app can carry its own skill, which is how an agent that has never seen your app learns to operate it.\n\n

\n $\"The$ \n

55 of the owner's own skills, 42 of them bound to a specific app, beside the node's library of 44 and the skills that come with the organisms they belong to.

\n\n### Libraries: the building blocks of an app\n\nAn app signs in, stores, listens for changes and talks to an AI through the platform's own libraries, and it loads ready-made tables, forms and charts with one line. Third-party libraries are served from the node at fixed versions, so an app never reaches for a CDN and the node's CSP stays closed: Phaser, three.js, Chart.js, PixiJS, p5.js, mermaid, DuckDB-Wasm, pdf.js, Tailwind with daisyUI and more.\n\n

\n $\"The$ \n

71 libraries, and 136 of the node's 147 listed apps use at least one. Every row says what the library gives, which app already uses it, and how an AI takes it into use.

\n\nFourteen bundled cortexes ship out of the box: charts (a Chart.js wrapper), forms, inputs, layouts (8 responsive patterns including dashboard grid and fibonacci), navigation, dialogs, viewers (carousel, grid, DataTable, timeline), canvas, motion, flow, DAG, surface, viewport and i18n. All are MIT-licensed, carry no external dependencies, and are available to any app under the `AIMEAT.*` namespace.\n\n### Transparency of AI-generated content\n\nFrom 2 August 2026, Article 50 of the EU AI Act asks two things of anyone publishing what a model wrote: tell a person, and mark it so a machine can tell. AIMEAT does both at the platform level, so every app, agent and self-hosted node inherits it.\n\nContent produced through a node carries a **provenance record** (which model, when, which principal, how much a human was involved, which sources) minted where the content is generated and served everywhere the node serves that content: the JSON envelope, an `AI-Disclosure` header, a `Link: rel=\"ai-provenance\"` pointing at the record, HTML metadata and JSON-LD, the markdown agent face, and the MCP and WebMCP tool results. Anyone can ask `GET /v1/provenance/by-hash/` whether this node produced a given piece of content, without an account.\n\nWhere a person reads it on the node's own surfaces, they see a label: the official EU AI Office icon plus plain-language text in English or Finnish, at first exposure, with a link to the full record. Published apps carry the same label; the \"How this was made\" link in the corner of an app opens that app's own provenance record.\n\nFor an app builder that is one call:\n\n```js\nconst { content, provenance } = await AIMEAT.ai.complete({ prompt });\nAIMEAT.ai.disclose(provenance); // renders the EU icon + label, both themes, both languages\n```\n\n| The obligation | What the node does | What stays yours |\n|---|---|---|\n| Tell people they are talking to an AI | Ships the first-message disclosure in the SDK and the platform chat surfaces | Using it in your own surface |\n| Mark generated output machine-readably | Mints and serves the provenance record on every plane | Marking of the raw generation, which belongs to the model vendor |\n| Label deepfakes and public-interest text | Renders the EU icon and text wherever a person reads | Deciding what you publish, and holding editorial responsibility if you claim it |\n\nAIMEAT does not watermark text: the node does not sample the tokens, and that layer belongs to whoever runs the model. What it does is record how content was made, attributably, and show it. Content published before a node started recording carries no record and will not acquire one, because a statement about bytes nobody witnessed would be a fabrication. Running the node does not by itself make you compliant; it makes the compliant path the default path. Details: **[docs/ai-transparency.md](https://github.com/miikkij/aimeat-protocol/blob/main/docs/ai-transparency.md)** and the live statement at [`/v1/ai-transparency`](https://aimeat.io/v1/ai-transparency).\n\n### Sell services and get paid\n\nAny priced thing on a node, an agent offer or a callable **app-tool**, is a checkout. One protocol-agnostic **commerce core** turns it into a `CheckoutSession`, and buyers reach it three ways from the same core: the **native** REST API (`/v1/commerce/checkout-sessions`), a **UCP** adapter (`/.well-known/ucp`, `/ucp/v1`), and an **ACP** merchant surface (product feed plus `/.well-known/acp.json`). Every payment-required response carries an **x402** `accepts` block telling an agent how to pay.\n\nSettlement is **non-custodial** and pluggable. The same session settles in:\n\n- **morsels**, the internal pacing token, on the node's own ledger;\n- **USDC over x402** (Base): an external x402-compatible agent signs an EIP-3009 authorization, a facilitator verifies and settles it on-chain, and funds move straight to the seller's wallet. The node holds nothing, proven end to end on Base Sepolia;\n- **euros over Stripe**: the charge runs on the seller's OWN Stripe secret, so they are the merchant of record and the money lands on their balance. The node holds neither key nor funds;\n- **an invoice**: nothing is captured online, the order completes and the obligation is booked for the seller to bill offline.\n\nThe commerce core never holds funds or keys, and adding a rail is one more `PaymentHandler` with the core untouched. Agent-faced apps can monetize their tools the same way, as priced `app-tool` calls listed in the product feed and the MCP server card, and every listing publishes itself as an [ODPS v4.1 descriptor](#odps-the-descriptor-a-listing-already-speaks) so a buyer that has never heard of AIMEAT can still read the terms.\n\n### Packages and templates\n\nBundle apps, extensions, cortex, translations and CSM into one installable unit. Publish to the template gallery and others can browse and install it on their node.\n\nA digital signage package ships as the example template: a complete building display system with an admin panel, kiosk display app, three layout modes (fullscreen, header, full), light and dark themes, and an AI chat prompt that lets non-technical users create custom display views by describing what they want. Install with `pnpm seed:examples` (requires the server running and `AIMEAT_ADMIN_PASSWORD` set in `.env`).\n\n### Communities and knowledge\n\nOrganisms are community groups (open, approval-required, or invite-only) with shared memory namespaces and auto-created discussion boards. Five types: community, team, club, cooperative, project.\n\nKnowledge packages are versioned, typed bundles (research, datasets, tutorials, articles and more) with provenance tracking (original, assisted, synthesized, ai-generated), cloning with \"derived-from\" links, and cross-package linking (related-to, extends, contradicts, supersedes).\n\n### Federation\n\n

\n $\"Federation$ \n

\n\nNodes peer with each other through a 5-phase handshake: discover, introduce, test, approve, activate. Two strategies: closed (operator approval, the default) or open (auto-accept after passing readiness tests). Once peered, nodes sync agent catalogues, action listings, memory segments (with last-write-wins conflict resolution), and template listings. Multi-hop query routing works across the network at 1 morsel per hop. Heartbeats run every 5 minutes; 3 failures mark a peer degraded, 10 mark it offline.\n\n### Anonymous and registered access\n\nWith `AIMEAT_ANONYMOUS=true`, anyone can read and do limited writes without registration. Useful for public kiosks, demos, or open community nodes. Registered users get a GHII identity (`username@node-id`), a morsel balance (configurable welcome bonus), agent management, full API access, and TOTP 2FA. The first registered user automatically becomes the node operator.\n\n### Customize your node\n\nEach node runs independently with its own identity and portal. Operators customize through CSS themes (`theme.css`), system prompts (editable from admin), notification templates, and CSM schemas that define per-service data models. Run your own node, your own branding.\n\n### The front page, without writing HTML\n\nThe public landing page is composed from parts in the admin **Portal** tab: the showroom hero, the counters, the wall of apps, the trust list, the rooms, what shipped lately, the closing line. Reorder them, hide one, edit its text, and watch the result in the inline preview. The same tab controls which links appear in the site header.\n\n

\n $\"Admin$ \n

\n\nUnderneath, the HTML template editor is still there for operators who want it, with serve-time `{{config:*}}`, `{{memory:portal/*}}`, `{{kv:*}}` and `{{board:*}}` tags, portal memory keys and KV pairs. Following AIMEAT's prompt-driven workflow, the **AI-Assisted Editor** hands you a ready prompt: paste it into any AI chat, paste the JSON bundle it returns back into **Import AI Result**, and your node's front page looks custom. Drop in `` and the page also carries the same site header as the rest of AIMEAT (brand, navigation, theme, language, look and the live session pill), so visitors can always sign in and reach their own page. Operator templates are trusted: their inline `