Navigating the UI

This page walks users through the GraphEngine Dashboard and Subgraph Querying userflow and UI.

GraphEngine Dashboard

The GraphEngine Dashboard provides a high-level view of platform activity. It is the first place users can check to see the health of the system, track how many subgraphs are live, and monitor usage across the network. The dashboard is designed to give both technical and non-technical users a quick snapshot of how GraphEngine is performing at any given time.

Active Subgraphs

Displays the current number of subgraphs that are live, accessible, and ready to be queried. This metric helps users gauge the scope of available data and provides a sense of how quickly new subgraphs are being added to the platform.

Pending Subgraph Requests

Shows the number of proposed subgraphs that are still awaiting review and approval. This gives visibility into the pipeline of new data structures being added and provides context on what might soon become available for querying.

Subgraph Queries

Tracks the total number of queries that have been processed across all active subgraphs. This highlights how the platform is being used in real time and reflects demand for specific datasets.

Chain Sync Status

Provides the current synchronization status of all chains supported by GraphEngine. Users can quickly see whether data ingestion from each network is up to date or if there are delays, ensuring confidence in the freshness and reliability of query results

Subgraph Page

The Subgraph Page provides a dedicated workspace for exploring and interacting with a specific subgraph. It is where users can see what data is being ingested, experiment with queries, and understand the underlying schema that powers their results. The page is designed for both casual users who want quick answers in plain English and advanced users who prefer to inspect raw configurations.

Playground

The chat-based playground allows you to query the subgraph directly using natural language. Simply type a question, and the system will generate a response by pulling structured data from the subgraph. This is the easiest way to interact without needing technical knowledge of query languages.

About

The About section provides details on the subgraph itself, including which events are being ingested and the specific on-chain contracts and wallets being tracked. This gives context on the scope of the subgraph so you know exactly what data your queries can touch.

Raw YAML

For advanced users, the Raw YAML section displays the exact YAML file that defines the subgraph. This shows which events, filters, and mappings are in place, giving full transparency into how data is captured and stored. It also serves as a reference for anyone interested in creating or modifying subgraphs in the future.

Chat History

The Chat History section organizes your past queries into threads, allowing you to revisit earlier questions, continue a line of investigation, or compare results over time. This makes it easier to build a continuous workflow rather than starting from scratch with every query.

Querying

The Querying UI is the primary way to interact with deployed subgraphs in GraphEngine. It allows you to experiment with natural language prompts, test outputs across different LLMs, and refine your approach with structured guidance. Whether you are a developer or a non-technical user, this interface is designed to make querying blockchain data intuitive and accessible.

Example Prompts

To help you get started, we provide a selection of pre-built example prompts. You can run these directly with a single click or use them as templates to design your own queries. Each example is aligned with the schema of the active subgraph for reliable results.

Choose LLM

When submitting a query, you can select which LLM model to use for processing. GraphEngine currently supports Claude Sonnet 4 and Artificial Super Intelligence’s ASI1 model. This flexibility allows you to experiment with different models depending on your needs for speed, accuracy, or cost efficiency.

Prompt Guidance

For best results, it is important to align your prompts with the structure of the subgraph you are querying. We’ve prepared a dedicated Prompt Guidance section that explains how to frame natural language queries effectively. We recommend reviewing this before creating complex queries.

Output

When you query a subgraph in GraphEngine, the system provides multiple outputs. Each serves a different purpose and gives users visibility into both the reasoning process and the raw data behind the final answer.

Cypher Query

The Neo4j Cypher query generated by the system in response to the user’s natural language input. This shows the exact database query executed against the subgraph, offering transparency into how the platform translated your intent into structured graph logic.

Natural Language

A plain language explanation generated from the Cypher query’s results. This provides a user-friendly summary of the data returned, making it easy to interpret insights without needing technical knowledge.

Data

The structured onchain data that underlies the response. This represents the actual blockchain events or entities pulled from the subgraph and used to form the natural language answer.

Raw JSON

The unformatted JSON payload returned by the query. This contains the complete data response from Neo4j, including entities, relationships, properties, and metadata. Developers and advanced users can use this for debugging, integration into external systems, or building custom visualizations.