About Aristotle

Aristotle is WattCarbon’s built-in assistant. It’s a read-only data analyst and platform guide, scoped to your own account: ask it about your assets, methodologies, and measurements in plain language, and it answers using your real data and the platform documentation.

What Aristotle can do

• Answer questions about your account — your assets, the methodologies applied to them, and their measurement results.
• Render tables and charts built from your account’s actual measurement data, so you can see a figure rather than read about it.
• Explain platform how-tos — onboarding meter data, configuring methodologies, and using the WattCarbon API — drawn from this documentation.
• Explain an asset’s electricity tariff — the rate plan it is billed on, including energy rates, fixed charges, demand charges, and the demand ratchet — and find candidate tariffs for an address.
• Analyze grid impact and non-wires alternatives — for a named distribution node or substation, Aristotle can pull grid-planning data (capacity constraint, DER potential, forward value streams, stress forecast, modeled load shape) and assemble a “defer the wire” scenario. These figures come from grid data rather than your account, and the forecast and load shape are modeled rather than metered.
• Help you onboard data — you can drop a spreadsheet or file into the chat; Aristotle analyzes it and, when it fits, proposes ingesting the data or creating sample assets for you to review.

What Aristotle can’t do

Aristotle’s data tools are strictly read-only, so on its own it cannot:

• create, edit, or configure assets or methodologies in your account — it can only propose a change (for example, ingesting a file you uploaded or creating sample assets), and the change happens only when you explicitly confirm it;
• change anything in your account or anyone else’s without that confirmation;
• compare your data against other customers’ — it only ever sees your account;
• answer off-topic questions. It declines requests outside WattCarbon’s domain (political advocacy, competitor analysis, regulatory or legal advice) and steers you back to your data.

If you need to change something, use the platform’s normal screens and workflows — Aristotle can tell you how, but it can’t do it for you.

Where the numbers come from

The figures in Aristotle’s tables and charts are computed from your account’s actual measurement data, then referenced by the assistant — they are not invented or estimated by the language model. Charts are built from the same query results that back the numbers in the narrative.

That said, Aristotle is an AI assistant and its responses can contain mistakes, especially in how it explains or frames a result. Always verify important figures against your dashboard and reports before acting on them.

Privacy and data handling

• Where it runs — your conversations are processed through AWS Bedrock using Anthropic’s Claude models.
• Retention — chat history is kept for 30 days, then automatically deleted. When an open conversation is within a week of that cutoff, the chat panel shows a banner naming the exact date it will be deleted.
• What’s logged — each turn records per-turn metadata and the read-only tool calls Aristotle made on your behalf to answer you. This supports the product, lets us audit answers, and helps prevent abuse.

Usage limits

Your account may have limits on Aristotle usage: a cap on messages per chat session, on new sessions per day, and on tokens per month. The per-session message and per-day session caps are unlimited by default. The monthly token allowance always applies — every account falls back to a platform default cap unless an administrator sets a different per-account value.

When you reach a limit, Aristotle tells you which one was hit — for example, that the per-session message limit or the monthly token limit has been reached. The per-day and per-month allowances reset automatically at the day and month boundary; the per-session limit resets when you start a new chat. If chat has been turned off for your account entirely, the panel says so.

Connect your own client

Aristotle’s chat is one way to use your data. You can also point your own MCP client (Claude Desktop, Cursor, and others) directly at WattCarbon’s read-only endpoint and use your own LLM token budget — see Connect your own client.