Integration

Neo4j Aura integration

Neo4j Aura is Neo4j's managed graph database cloud. With the Aura API an agent can list projects and their allowed instance configurations, create, resize, rename, pause, resume, upgrade, overwrite and delete AuraDB and AuraDS instances, take, list and restore snapshots, estimate instance and Graph Data Science session sizes, manage GDS sessions and customer managed encryption keys, and read a project's metrics integration endpoint. Running Cypher queries against a database is not part of this API, and the database password Aura generates for a new instance is not passed on. Authentication is OAuth 2.0 client credentials: create an Aura API client id and secret in the Aura console (Account settings, Client credentials); one-hour access tokens are requested from the token endpoint with HTTP Basic auth.

Connect Neo4j Aura How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Neo4j Aura.Built from the 26 Neo4j Aura tools below, on the connection you add once.

What you can do with Neo4j Aura

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Instance, 5 operations

create_instanceCreate instancePOSTCreates a billed Aura instance (async: poll get_instance until running) from one of get_project's instance_configurations. The generated password is not returned: a project admin sets one in Aura console Query (CREATE USER ... SET PASSWORD).
delete_instanceDelete instanceDELETEDeletes an instance and its data permanently (asynchronous; the instance shows status destroying until gone).
get_instanceGet instanceGETReturns one instance: status (creating, running, paused, updating...), connection_url, type, region, memory, storage, version options and flags. Poll this after an asynchronous instance operation.
list_instancesList instancesGETLists the Aura instances (id, name, project, cloud provider, creation time), optionally for one project. Use get_instance for status and size.
update_instanceUpdate instancePATCHRenames, resizes (memory, storage) or toggles vector optimization or Graph Analytics on an instance. Send only what changes. Asynchronous, the instance stays available: poll get_instance until status is running.

Customer managed key, 4 operations

create_customer_managed_keyCreate customer managed keyPOSTRegisters a key from your cloud provider's KMS in Aura (asynchronous). Before use, grant Aura permissions on it as shown in the Aura console under Customer Managed Keys.
delete_customer_managed_keyDelete customer managed keyDELETEDeletes a customer managed key from Aura. Refused (HTTP 400) while any instance uses it.
get_customer_managed_keyGet customer managed keyGETReturns one customer managed key: cloud provider, region, instance type, the key id in your cloud, status and project.
list_customer_managed_keysList customer managed keysGETLists the customer managed encryption keys (id, name, project), optionally for one project.

Session, 4 operations

delete_gds_sessionDelete gds sessionDELETEDeletes a Graph Data Science session, discarding its in-memory graphs.
get_gds_sessionGet gds sessionGETReturns one Graph Data Science session: status (Creating, Ready, Expired or Failed), host, memory, ttl and expiry.
get_or_create_gds_sessionGet or create gds sessionPOSTReturns the GDS session with this name in the project, or creates it (billed compute, by memory and time). Attach it to instance_id, or make it standalone with cloud_provider and region. Poll get_gds_session until Ready.
list_gds_sessionsList gds sessionsGETLists Graph Data Science sessions (name, memory, status, attached instance, expiry), optionally filtered by instance, project or organization.

Snapshot, 3 operations

create_snapshotCreate snapshotPOSTTakes an on-demand snapshot of an instance (asynchronous). Returns snapshot_id; poll get_snapshot until status is Completed.
get_snapshotGet snapshotGETReturns one snapshot: profile (AdHoc or Scheduled), status (Completed, InProgress, Failed, Pending or Cancelled), timestamp and whether it is exportable.
list_snapshotsList snapshotsGETLists an instance's snapshots for one day (today by default): snapshot_id, profile (AdHoc or Scheduled), status and timestamp.

Sizing, 2 operations

estimate_gds_session_sizeEstimate gds session sizePOSTEstimates the memory a GDS session needs for a given graph, without creating anything: recommended_size and estimated_memory.
estimate_instance_sizeEstimate instance sizePOSTEstimates the memory an instance needs for a given graph size, without creating anything: recommended_size, min_required_memory, did_exceed_maximum.

Tenant, 2 operations

get_projectGet projectGETReturns one project with its instance_configurations: every allowed combination of cloud_provider, region, type, version, memory and storage. Read it before create_instance or update_instance.
list_projectsList projectsGETLists the Aura projects (formerly tenants) the credentials can reach, with id and name. A project id is the tenant_id/tenantId other tools take.

Metrics integration, 1 operation

get_project_metrics_integrationGet project metrics integrationGETReturns the metrics integration endpoint URL of a project.

Overwrite, 1 operation

overwrite_instanceOverwrite instancePOSTReplaces ALL data of the target instance with a copy of source_instance_id (a new snapshot of it, or source_snapshot_id). Irreversible. Asynchronous: poll get_instance until status is running.

Pause, 1 operation

pause_instancePause instancePOSTPauses an instance (asynchronous; status pausing, then paused). A paused instance cannot be queried until resumed.

Restore, 1 operation

restore_instance_snapshotRestore instance snapshotPOSTRestores an instance to a snapshot, replacing its current data. Irreversible. Asynchronous: poll get_instance until status is running.

Resume, 1 operation

resume_instanceResume instancePOSTResumes a paused instance (asynchronous; status resuming, then running): poll get_instance until status is running.

Upgrade, 1 operation

upgrade_instanceUpgrade instancePOSTUpgrades an AuraDB Professional instance to Business Critical (billed at that tier). Not for Marketplace projects or trials. Asynchronous: poll get_instance.

Use Neo4j Aura from Claude Code, Cursor or Codex

Connect Neo4j Aura in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Neo4j Aura operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Neo4j Aura

Create your workspace
Sign up free. Nothing to install.
Connect Neo4j Aura once
Add your Neo4j Aura connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.