Skip to content

PuppyGraph Web UI

PuppyGraph provides a web-based UI for graph modeling, querying, visualization, and cluster management.

This guide introduces the main pages of the Web UI. For programmatic access, see PuppyGraph CLI and Client Drivers.

To use the UI, make sure your browser meets the System Requirements.

Login

The login page asks for a username and password.

The default username is puppygraph. The default password depends on how the instance was launched:

  • The value of the PUPPYGRAPH_PASSWORD environment variable for instances started with Docker.
  • The EC2 Instance ID for instances launched from the PuppyGraph Professional AMI in AWS Marketplace.
  • The Compute Engine Instance ID for instances launched from the PuppyGraph Professional VM in Google Cloud Marketplace.
PuppyGraph Sign-In Page
PuppyGraph Sign-In Page

Left navigation

After signing in, the sidebar exposes the main pages:

Page Purpose
Graph Build, view, and edit the graph schema.
Catalogs Manage catalog connections and locally cached tables.
Chat Use the built-in AI chatbot. Appears when AI Chat is enabled.
Query Run Cypher or Gremlin queries against the graph.
Dashboard Customizable tiles for counts, sample data, and saved queries.
Cluster Cluster health and per-node status.
Settings Account preferences, RBAC administration, AI Chat configuration, and server controls.
Help Links to documentation and product help.

Graph

The Graph page is the home of the Graph Modeling workflow. On a fresh instance, the page shows three quick-start tiles:

  • Connect Data Source: open the Schema Builder and add your first catalog.
  • Use Example Graph: load the built-in TinkerPop modern graph (or the Northwind graph).
  • Upload Schema: upload a v1 schema JSON file directly.
Graph welcome screen
Graph welcome screen

Once a schema is loaded, the page shows the schema canvas with the modeled nodes and edges. The toolbar above the canvas exposes the editing affordances:

  • Download and Upload export and import the schema JSON.
  • Reset clears the schema.
  • History lists prior schema versions.
  • Add Node and Add Edge open the Schema Builder wizards.
  • Nodes, Edges, and Local filter what the canvas highlights.
Graph page with a loaded schema
Graph page with the modern graph loaded

For the full Schema Builder walkthrough, see Building a Graph and the Schema Builder tutorial. For uploading and downloading the schema JSON, see Managing the Graph.

Catalogs

The Catalogs page lists every catalog connection on the instance and lets you inspect and manage them.

Catalogs page
Catalogs page with no catalogs configured

From this page you can:

  • Add a new catalog connection (Add Catalog) without leaving the page.
  • Refresh the list and the underlying database / table metadata.
  • For an existing catalog, drill into its databases and tables.
  • Manage locally cached tables (status, refresh, partition details, source info, import history).

See Data Sources and Local Tables for the underlying configuration.

Chat

The Chat page contains PuppyGraph's built-in AI chatbot. It appears in the left navigation when AI Chat is enabled.

Use Chat to:

  • Ask PuppyGraph to inspect connected catalogs and build a graph schema.
  • Generate and run graph queries from natural-language questions.
  • Explain query results using the active graph schema as context.

For setup, usage patterns, and operational guidance, see Built-in AI Chatbot. For a complete schema-building walkthrough, see Building a Graph with the AI Assistant.

Query

The Query page is a single editor panel that runs Cypher or Gremlin queries against the current graph. The language is toggled from the toolbar (the Cypher button cycles between Cypher and Gremlin), and the timeout is configured next to it.

Below the editor are tabbed result views (Graph, Table, Schema, Raw) and a history panel of prior runs.

Query page
Query page in Cypher mode, with the schema visualizer in the lower panel

See Querying for query-language details. The result panel's Graph tab handles the visualization; the Dashboard page is the place to assemble persistent saved views.

Dashboard

The Dashboard page is a customizable grid of tiles for sharing high-level views of the graph: counts of nodes / edges by label, sample rows, embedded query results, and a graph-samples panel.

PuppyGraph Dashboard
Dashboard with built-in Node / Edge counters, Node Labels breakdown, and a Graph Samples panel

See Dashboard for details.

Cluster

The Cluster page shows the cluster topology as two sections of node cards. The header shows the overall cluster status and the active schema version.

Cluster page
Cluster page showing leader and compute nodes with their roles, connection state, and resource usage

Each Leader card shows the node's address, role in the quorum, and service health.

Each Compute card shows the node's address, service health, and current resource usage.

For how to deploy a PuppyGraph cluster and manage its nodes, see Cluster Deployment.

Settings

The Settings page is organized as tabs:

Tab Purpose
Preferences UI Preferences (e.g. disabling the graph-canvas render to reduce GPU usage), the AI Chat API key, password change, and server management (restart, download logs).
Users Local user accounts and role assignments. Available when RBAC is enabled.
Service Accounts Programmatic API credentials. Available when RBAC is enabled.
SSO Configuration Configure SSO providers (OIDC, SAML). Available when RBAC is enabled.
Audit Log View administrative actions. Available when RBAC is enabled.
Settings page
Settings → Preferences with UI, AI Chat, Password, and Server Management sections