Hello world. Chartnaut is the ultimate agentic trading environment for the modern trader — and this is the first post on the blog.
We are building one workspace where you can practise on history, trade live markets, research with AI across time and data dimensions, and carry what you learn back onto the chart when it matters. Not a charting app with a chatbot bolted on. Not a black-box bot that trades for you. A platform that makes the trader sharper.
The vision
Markets are getting denser, faster, and more instrument-rich. The edge will not belong to whoever has the flashiest indicator pack. It will belong to individuals who can ask better questions of more data — and act on the answers without leaving the terminal.
Chartnaut’s vision is to support every asset class you care about, and to let AI research across time on every dimension of market data we can wire in: candles, ticks, MBO / order-book depth, volume tape, and the layers that come next. One environment. Many markets. Research that can actually see the microstructure, not just the close.
Our thesis is simple: the future of markets is defined by individuals powered by agents — not by fully autonomous agentic trading that removes the human from the loop. We want deeper tooling, more tightly integrated agents, and a stack that helps you build, measure, and execute with evidence. Forward Insights on the chart. Definitions and studies that speak your language. A journal that becomes a queryable engine. This is just the start.
What we are not building
We are not building a set-and-forget autopilot. Autonomous agents that place risk without you in the seat sound exciting until the drawdown is yours. Chartnaut’s agents draft indicators, author definitions, run studies, search your journal, and put measured answers on the chart — so you decide with better context.
Empowerment over autonomy. Tooling over mystique. That is the bet.
The core of the platform
Five pillars hold the product together today. Each one is already shipping — and each one is a foundation for what comes next across asset classes and data types.
1. Simulate through backtesting sessions
Backtesting in Chartnaut is not a spreadsheet of hypothetical fills. It is a session in the same Terminal you use live: replay historical markets bar by bar, practise the setup, pause, resume, and file trades as you go.
Practice stays honest because the workspace is honest — same charts, same order confirmations, same snapshots. When a session holds up, you can promote what you learned into a playbook instead of rebuilding the idea by hand.
2. Real-time trading
Research that never reaches execution is a notebook. Chartnaut connects Demo and MetaTrader into the same Real-time Terminal mode you use for replay — live clock, live prices, positions and orders on screen.
The point is continuity. You should not practise in one app, journal in another, and trade in a third. Capture, review, and execute from one place so the feedback loop stays closed.
3. Agents and deeply integrated agentic tools
Agents in Chartnaut are not a side panel that “chats about trading.” They are wired into the objects you actually use:
- Indicators — describe a plot, preview it, save it, reuse it in live and practice
- Definitions — name a market situation so Chartnaut can find every occurrence in history
- Studies — ask what usually happens after that situation, measured across the sample
- Forward Insights — put a finished study on the chart so the answer shows up when the setup appears
That loop — define, collect, study, pin — is how agentic tooling should feel: you stay in control, the agent collapses the busywork, and the evidence lands where you decide.
4. Your trade and account data as a queryable engine
Every fill that lands through Chartnaut — live, demo, sim, or replay — becomes part of a living dataset: the trade library, account reports, tags, notes, snapshots, playbook assignments.
We treat that data as something you can continually query and study. Trade studies let you ask plain-English questions about your own journal — “do I trade better away or closer to VWAP?” — and get measured answers from real rows, not a paragraph guessing what happened.
Your history should compound. The more cleanly you file and tag, the sharper the engine gets — for weekly review, for playbook cleanup, for the next live decision.
5. Playbooks and Historical Intelligence
A playbook is the ideal version of one strategy: rules, graded examples, filed trades, and outcomes kept separate from the rest of your account noise.
Historical Intelligence trains on those examples — per playbook, per market — so Chartnaut can surface likelihoods grounded in how you have traded that setup, not a generic signal from the internet.
Build the book honestly. Collect a real sample. Then let Historical Intelligence sit beside live and practice decisions as a guardrail — context when you are about to take the same setup again.
Where this is going
Today Chartnaut already closes a serious loop: simulate, trade, research with agents, study your journal, and guide live decisions with playbook memory. Ahead of us: broader asset coverage, deeper market-data dimensions, and agents that reach further into the workflow without ever taking the seat from you.
The blog is where we will write about that journey — shipping notes, research workflows, and the thinking behind the product. Feature deep-dives live under Features. This space is for the story around them.
Start here
Skim the full feature stack, start free, and file a few honest trades to one playbook. Once the record exists, backtesting, studies, trade studies, and Historical Intelligence all have something real to work with.
Welcome aboard. This is just the start.
