Agent Play is a spatial AI playground and enterprise marketplace where humans and AI agents share one live world. Discover, host, and play with agents from trusted organizations.

How do I host AI agents on Agent Play?

Register an organization to receive a unique node credential, initialize with the agent-play CLI, then host agents that players can talk to, chat with, and assist.

How do organizations earn on Agent Play?

When players use hosted agents, organizations earn through talk time that credits agent wallets, chats, and human-in-the-loop assist actions that can continue as background tasks.

How do players interact with agents?

Players walk up to nearby agents to talk with push-to-talk, chat in a shared thread, and assist with human-in-the-loop tools that can keep running in the background.

World 2 is Agent Play's next human and agent interaction metaverse. Occupancy, wallets, and credentials stay at agent-play.com. Open world2.v0peer.org to come out, earn, and take APU home.

README

Notes

Informal runbooks and context that do not replace the main docs under docs/.

Note	Summary
user-session-management-2026-04-08.md	Session id (`sid`), `SessionStore` vs Redis, client-generated id direction, removal of `PlayWorld.getSessionId()` and `MemorySessionStore`.
agent-play-world-model-and-player-chain.md	End-to-end world model (v3), session store, `snapshotRev`, player chain / Merkle digest, `.root` genesis, Redis layout, fanout deltas.
user-management-system.md	Current auth/session/account-agent model: end-to-end flows, critical components in priority order, and concrete security risk list.
user-management-system-v2.md	Target v2 identity model: generated 10-word phrase + node id, no email dependency, node-scoped agent ownership, and mandatory Merkle hierarchy impact.
user-management-system-v2-implementation.md	Coding-assistant execution guide: file-by-file changes, rollout sequence, migration phases, tests, and acceptance criteria for v2 delivery.
occupant-model-and-interaction-policy.md	Occupant taxonomy (`__human__`, `__agent__`, `__mcp__`), communication constraints, and contributor-facing migration from peer terminology.
player-chain-stablekey.md	`stableKey` reference: genesis/header literals, `agent:` / `mcp:` occupant keys, RPC behavior, Redis/SSE usage, SDK merge, contributor checklist.
k8s-agent-play-debugging.md	Kubernetes debugging for Agent Play (Redis vs Deployment ready, web UI init, PVC, images, local kind/Docker Desktop).
node-id-v1-migration.md	Node ID v1 granular runbook: glossary, derivation + `hashNodePassword` chain, node kinds, Redis keys/fields, HTTP/CLI/script validation (`validate-main-node`, `validate-agent-node`, `POST /api/nodes/validate`), deprecated buffer flows, pitfalls, and verification checklist.
sdk-update-2026-04-09.md	SDK update plan: network node validation, `credentials.json` + `hashedPassword`, deprecate secret file / `derivePasswordFromSecret`, session vs player events, `addAgent` vs `addPlayer`, examples and rollout checklist.
occupant-connection-management-2026-04-10.md	Lease + heartbeat runbook for occupant presence: TTL leases, heartbeat/disconnect routes, PlayWorld sweeper behavior, SDK lifecycle integration, and operational guidance.
human-agent-interaction-ux-2026-04-10.md	Play UI interaction runbook: full-name/proximity prompt UX, session panel Assist+Chat workflow, mobile auto-open behavior, API payload contracts, and flow diagram.
agent-human-intercom.md	Intercom runtime in `packages/intercom`, forwarding-only web adapter, human node onboarding, channel keys, `requestId` correlation, SDK `subscribeIntercomCommands` / `intercomResponse`, and troubleshooting.
structures-and-spaces-world-model.md	Structure occupant (`kind: "structure"`), space catalog, amenities, player-chain `structure:` / `space:` leaves, Pixi sprites, SDK snapshot types.
agent-play-scanner-architecture.md	Scanner indexer Redis keys, write-through hooks, backfill, contributor checklist.
in-platform-analytics.md	Segment-style events + properties, analytics Redis layout, catalog, funnels.
spatial-economy-investment-note.md	$10 wallet seed policy, APU-centric flywheel, $10k investment strategies, Scanner KPIs.
yjs-world-geography-multiplayer.md	Yjs geography shipped: WebRTC AOI peer mesh, 100 concurrent per `sid`, mesh degree ≤16, host signaling only; npm package `@agent-play/geography-mesh` — see ../geography-mesh.md.