SCADA Miner NEM Explorer is an AI-powered tool for querying Australia’s National Electricity Market (NEM). It helps analysts, developers and market participants ask plain-English questions about AEMO data and receive answers backed by SQL queries, charts and tables.

This FAQ explains what NEM Explorer can answer, where the data comes from, how credits work and when to use the chat interface versus connecting using the NEM Explorer MCP server using an MCP-compatible agent such as Claude Code, Cursor, Codex, Windsurf, VS Code Copilot, Gemini CLI, Zed, Continue, Cline or Roo Code.

Open NEM Explorer or contact SCADA Miner if you need help with a use case that is not covered below.

Getting started with NEM Explorer

What is NEM Explorer?

NEM Explorer is an AI interface for Australia’s National Electricity Market data. You can ask questions in plain English and get answers based on real AEMO data, including generated SQL, charts and tables. You do not need to write SQL or code to use it.  Two methods of analysis are supported

  1. A chat interface using a low-cost medium-intelligence AI. Use this to understand the available data and get answers to questions about the data.
  2. An MCP server. Use this to connect your own agent (e.g. Claude Code Opus or Codex) for much more powerful analysis.

Both options make validated capabilities available to the agent, so when asking a question like “how much money did Hornsdale BESS make on 1 April?” the agent uses a pre-written and validated SQL query incorporating things like linear ramping between dispatch targets, and income from FCAS as well as energy.

How do I ask a question?

Either type your question into the chat box or connect your agent to the MCP server.

  • What was the average spot price in Victoria last week?
  • Compare wind and solar generation in South Australia over the past month.
  • Show interconnector flows between New South Wales and Queensland for the last seven days.

What kinds of NEM questions can I ask?

NEM Explorer is designed for questions about the wholesale electricity market and any data AEMO hosts on nemweb, including:

  • Spot prices, negative prices and price spikes by region.
  • Generation output by fuel type, power station or dispatchable unit identifier (DUID).
  • Interconnector flows and constraints.
  • FCAS, frequency control markets and enablement.
  • Demand forecasts and pre-dispatch data.
  • Bidding and offer behaviour.
  • Capacity factors and generation mix trends.

Chat answers versus connected AI agents

What does the chat interface answer well?

The chat interface works best for one-shot answers. You ask a concrete question, NEM Explorer selects the relevant table or tables, runs a SQL query and writes the answer. It is fast and consistent for:

  • Factual lookups, such as the spot price for NSW1 on a specific date.
  • Canonical rankings, such as the top batteries by revenue in a quarter.
  • Standard aggregations, such as fuel mix for QLD over the last 30 days.
  • Questions that resolve to specific DUIDs, regions and intervals.

When should I use an MCP-compatible agent instead of chat?

Some market analysis questions need an agentic workflow rather than a single chat response. In those cases, connect an MCP-compatible agent to NEM Explorer so the agent can inspect tables, refine queries and follow the analytical trail across multiple steps.

Use a connected agent for:

  • Open-ended exploration, such as investigating battery profitability and identifying what is interesting.
  • Multi-step research, such as finding each constraint that bound and then estimating affected DUIDs and lost revenue.
  • Pattern or anomaly detection, such as finding when unit output deviates from typical behaviour.
  • Counterfactual modelling, such as estimating what may have happened if a constraint or unit was removed.
  • Compound questions with several distinct sub-questions.

When the chat detects that a question needs this kind of workflow, it can redirect you to the agent workflow instead of forcing a shallow single-shot answer.

What does “Connect an agent” do?

The Connect an agent menu opens an installer that wires the NEM Explorer MCP server into your agent of choice. Supported tools include Claude Code, Cursor, Codex, Windsurf, VS Code Copilot, Gemini CLI, Zed, Continue, Cline and Roo Code.

The agent receives tools for querying the NEM Explorer warehouse, including search_entities, get_data_freshness, list_tables, get_table_schema, search_columns, list_capabilities, run_capability and execute_sql. Authentication is handled with your existing NEM Explorer account. Query results are capped at 500 rows, so the agent paginates or aggregates as needed.

Claude Code users can also install the SCADA Miner plugin with /plugin marketplace add brensch/scadaminer-ai-toolkit followed by /plugin install scadaminer@scadaminer. The plugin adds the same connector plus NEM slash commands and an analyst subagent.

AEMO data sources and NEM coverage

Where does the data come from?

NEM Explorer uses data from AEMO’s NEMweb, the official public data portal for the National Electricity Market. This includes dispatch data, SCADA readings, trading intervals, bids and offers, constraints, interconnector flows, FCAS markets, demand forecasts and other market datasets.

Which NEM regions are covered?

NEM Explorer covers all five NEM regions: New South Wales (NSW1), Victoria (VIC1), Queensland (QLD1), South Australia (SA1) and Tasmania (TAS1).

What time resolution is available?

The NEM operates on 5-minute dispatch intervals and 30-minute trading intervals. Most dispatch, generation and SCADA data is available at 5-minute resolution. Some datasets, including trading prices and PASA-style datasets, are available at 30-minute or daily resolution.

When I ask about a “day”, is that a calendar day or a trading day?

NEM Explorer treats dates as calendar days in NEM time, which is Australian Eastern Standard Time (UTC+10). NEM time does not observe daylight saving. It remains a fixed UTC+10 offset year-round, so questions about “yesterday” or “last Tuesday” use the NEM calendar date rather than UTC or your local timezone.

How far back does the data go?

Data availability varies by dataset. Core dispatch and pricing data typically goes back several years. NEM Explorer tracks the available date range for each table and will tell you when a question falls outside the available window.

Accuracy, limitations and interpretation

How accurate are the answers?

NEM Explorer generates SQL queries against AEMO data, so the underlying market data is as accurate as AEMO’s public publications. The AI layer that translates a question into SQL is tested against sources such as the Open Electricity API, AER published figures and other reliable references.

As with any AI system, there is still a possibility of misinterpretation. Always check whether the generated query, filters and time range match your analytical intent.  For information about validation, refer to this post on external validation sources.

Is rooftop solar included?

Rooftop solar and other behind-the-meter distributed generation are not centrally metered by AEMO and are usually not included by default as direct generation data in NEM Explorer. The generation data focuses on utility-scale generators that participate in the wholesale market. In most cases, where there is potential for ambiguity, a note is added to the response clarifying how rooftop solar is treated in the answer.

Why might generation totals differ from other sources?

NEM Explorer uses dispatch data, including TOTALCLEARED and INITIALMW, to estimate energy production with a trapezoidal approximation across 5-minute intervals. Other sources may use metered data, which is published later and can differ slightly.

Some smaller solar and wind farms report generation through SCADA only intermittently or not at all. For aggregate generation questions, dispatch data generally provides more complete coverage than raw SCADA readings.

Does NEM Explorer include distribution network or DNSP data?

NEM Explorer covers wholesale market data only. Distribution network data, including local usage patterns, demand charges and DNSP-specific tariffs, is not currently included. If you need DNSP data, contact SCADA Miner to discuss your use case.

Does NEM Explorer include retail electricity pricing?

Retail pricing, tariff structures and demand charges are not currently available. NEM Explorer focuses on the wholesale spot market and related AEMO datasets. If retail pricing is important to your workflow, get in touch.

Pricing, credits and usage

How much does NEM Explorer cost?

NEM Explorer uses a credit-based system. Users can view their current credit packs and pricing from the Usage page in the chat application by selecting Buy credits.

How many credits does a question use?

Credit usage depends on query complexity. A simple price lookup may use around 50 credits. A more complex analysis that joins several datasets may use 200 credits or more. The credit cost is shown after each question.

Are credits shared between chat and connected agents?

Yes. Your account has one credit balance for both ways of using NEM Explorer:

  • Answers generated in the chat interface.
  • Data queries run by your connected agent over MCP, including calls to execute_sql or run_capability.

Both draw from the same credit pool, so there is nothing separate to top up for agent access.  Try the MCP server for complex research on the AEMO data.

Is there a free allowance?

Plans can include a free monthly allowance that refreshes at the start of each month. The free allowance is non-accruing, so unused free credits do not roll over and expire at month-end. Your current allowance and expiry are shown on the Usage page.

Do credits expire?

Free monthly allowance credits expire at the end of the month in which they were granted. Purchased credits remain valid until their own expiry, which depends on your account plan. Check the Usage page for your current balance, free allowance and expiry.

Feature requests, data exports and API access

How do I request a feature?

Visit the SCADA Miner contact page to submit a feature request. Common requests include DNSP data integration, data exports and API access.

Can I export data or get API access?

Bulk data export and API access are not currently available through the chat interface. If you need programmatic access or large-scale data extraction, contact SCADA Miner to discuss your requirements.

Start using NEM Explorer

To ask questions about AEMO data, NEM spot prices, FCAS markets, generation output, constraints or interconnector flows, open NEM Explorer.