Memory Cloud
Memory Cloud is the same platform as a self-hosted install, run and upgraded by us. There is nothing to install and nothing to bootstrap: sign in, and the next three steps are a project, a blueprint and a client.
1. Sign in
Section titled “1. Sign in”Open app.memory.emergent-company.ai and select Continue with your Memory account. Sign-in goes through our identity provider, so it is one account across the UI, the CLI and the API.
2. Open a project
Section titled “2. Open a project”A project is the workspace everything else lives in: graph objects, documents, chat sessions, agents, branches. Your organization may already have projects, in which case they are listed for you.
If it does not, select New project, give it a name and — optionally — a purpose: one sentence describing what knowledge it holds. The purpose is passed to AI features as context, so a real sentence beats a label.
3. Install a blueprint
Section titled “3. Install a blueprint”An empty project has no object types, so the first thing to add is a domain model. Install a blueprint rather than building one by hand: it bundles object and relationship types, agent definitions, skills and seed data and applies them in one action.
- Open the project and go to Blueprints.
- Pick one from the registry.
task-boardis a good first install: a board-enabledTasktype, a task-worker agent with its workflow skill, and one example task. - Select Install. The blueprint’s types appear in the schema, its seed objects land in the graph, and the badge on the blueprint row reads Installed.
Everything a blueprint created is owned by that blueprint, so Uninstall removes exactly what it added. Built-in blueprints are immutable — fork one with New version to change it.
4. Connect a client
Section titled “4. Connect a client”You now have a project with a working model. Three ways in, same credentials, and you only need one of them.
Every option authenticates as the same project and reaches the same graph, so pick the one that fits where the client runs: an MCP client reads and writes the graph as tools, the CLI is for a terminal or a script, and the REST API is for anything else.
Memory is itself an MCP server, so an AI client — Claude, Cursor, OpenCode — reads the graph directly, as tools, with no code in between.
- Open the project in Memory (
app.memory.emergent-company.ai) and go to MCP Sharing. - Select Create a share, name the recipient and choose read-only or read-write. Memory mints a token scoped to that share.
- Copy the install link or the JSON config the dialog shows. The link is private — it carries the token.
- Paste the config into the client's MCP settings and restart it. The client lists the
memory_*tools and can search the graph.
{ "mcpServers": { "memory": { "type": "http", "url": "https://api.memory.emergent-company.ai/api/mcp", "headers": { "X-Project-ID": "<project-id>", "X-API-Key": "<scoped-token>" } } }}# 1. install the CLI (adds ~/.memory/bin to PATH)curl -fsSL https://raw.githubusercontent.com/emergent-company/emergent.memory/main/deploy/install-cli.sh | bash
# 2. sign in to this instance — the browser opens, the session is savedmemory login --server https://api.memory.emergent-company.ai
# 3. check what the CLI can see, then ask the graph a questionmemory projects listmemory query "what have we decided about pricing?"memory login opens the browser and stores an OAuth session, which is the right choice when one sign-in should cover several projects. For CI, skip the browser with memory set-token <token> or a MEMORY_PROJECT_TOKEN environment variable — theCLI referencehas the full credential resolution order.
Mint a token in Project settings → API tokens → Create token, or from the CLI withmemory projects create-token --project <name>. The token is scoped to the project and shown once.
# the token is shown once, at creation — keep it in an environment variableexport MEMORY_TOKEN=emt_...export MEMORY_PROJECT=<project-id>
curl -s https://api.memory.emergent-company.ai/api/blueprints \ -H "Authorization: Bearer $MEMORY_TOKEN" \ -H "X-Project-ID: $MEMORY_PROJECT"Send X-Project-ID on every project-scoped call;Authorization: Bearer and X-API-Key are both accepted. The same credentials reach the MCP endpoint athttps://api.memory.emergent-company.ai/api/mcp.
What’s next
Section titled “What’s next”- Knowledge Graph — objects, relationships, search and traversal.
- Agents — scheduled and event-driven work over the graph.
- Data Sources — pull external systems in and keep them in sync.
- Self-hosted — the same platform on infrastructure you control, for data residency or a private network.