Questions traders actually ask

Active traders who want to combine charting, AI analysis, backtesting, automation, and journal review in one integrated tool — rather than managing five separate applications that don't talk to each other.

Dome Terminal is equally useful for discretionary traders (who use ORACLE, Quant Brain, and the journal to sharpen their edge), systematic traders (who use Python Lab, Script Library, and Bot Management to build and deploy rule-based strategies), and prop firm traders (who need strict risk gate controls, detailed journal records, and drawdown management built into their workflow).

If you want AI to generate signals for you and you don't want to think about it, Dome Terminal is not the right tool. It is built for traders who want more context, more discipline, and more insight into their own decision-making — not for traders who want to outsource those decisions entirely.

No Dome Terminal does not provide trading signals, guaranteed calls, or trade recommendations.

The terminal provides context, analytical tools, and AI assistance — Quant Brain readings, ORACLE analysis, AI Committee deliberations, backtest results, and journal insights. These are inputs to your decision-making process, not substitutes for it. Every trade decision remains with you.

Any AI output from Dome Terminal is advisory and probabilistic — not a guaranteed outcome. If you're looking for a service that tells you what to trade, this is not it. If you're looking for a system that helps you make better-informed decisions yourself, this is built for you.

Partially. Dome Terminal is designed to be local-first, which means a significant portion of the product works without an internet connection:

• Available offline: Charts from cache (the most recent data that was loaded while online), Trade Journal (all entry, review, and tagging functions), Script Library editing, Python Lab notebooks, and local AI models (if downloaded)
• Requires internet: Live market data, remote AI providers (like Claude or GPT), news feeds, Market Intelligence live data, and account sync

For traders in areas with unreliable internet, the combination of cached data + local AI + offline journal means the core research and review workflow remains functional even when the connection drops.

Yes Dome Terminal is built on a local-first architecture. By default, your journal entries, trading strategies, script library, cached market data, Python Lab notebooks, and AI conversation history all stay on your device — they are not uploaded to our servers.

The one area where this requires active attention is remote AI context routing: when you use a remote AI provider (rather than a local model), the prompt and selected context is sent to that provider's API. By default, sensitive fields like account balances, live PnL, and strategy scripts are excluded from remote AI context. You control exactly what is included through Settings → AI → Privacy.

If complete data privacy is a requirement, use local AI models (Ollama or built-in GGUF) and nothing leaves your machine at all.

Yes All three asset classes are supported, subject to what your configured data providers cover.

• Crypto: Spot and perpetual futures pairs on major exchanges (Binance, Bybit, OKX, Coinbase, and others depending on provider)
• Equities: US stocks and ETFs through supported equity data providers. International equities depend on provider coverage.
• Futures: Exchange-traded futures (ES, NQ, CL, GC, and others) through CME and other supported futures data providers
• Forex: Major and minor pairs through connected forex data providers

The specific symbols and timeframes available depend on which data providers you have configured and what your subscription tier with those providers includes. Check Settings → Data Providers to see your active provider coverage.

ORACLE is a context-aware AI assistant that reads your active terminal state before responding to any question. Before answering, it reads: the symbol and price on your active chart, current Quant Brain readings (trend, volatility, structure, delta), your most recent journal entries and mistake tags, your current portfolio positions, and your active risk configuration.

A regular AI chatbot has none of this. It responds with generic market commentary because it doesn't know what you're looking at, what your position is, or what your journal says about your behavioral patterns. ORACLE's value comes from the specificity it can offer because of the terminal context — which is also why asking specific, grounded questions produces much better results than asking vague questions.

Use ORACLE for: market state analysis, challenging your trade thesis, reviewing your journal patterns, and sense-checking position sizing relative to current conditions. Not for: guaranteed calls, execution decisions, or outsourcing your judgment.

No ORACLE is a read-only analytical tool. It has no ability to place orders, modify positions, change settings, or interact with any execution system.

This is a deliberate design decision. AI analysis and trade execution are intentionally separate layers in Dome Terminal. ORACLE provides analysis and challenge. You decide what to do with that analysis. Execution actions — whether manual trades or bot activations — require explicit trader input. There is no pathway from ORACLE output to automatic execution.

The AI Committee runs 8 specialist AI agents simultaneously against a trade proposal you submit. Each specialist examines the idea from a different angle: Technical (price structure), Orderflow (delta and depth), Sentiment (news and social signals), Risk (position sizing and drawdown), On-Chain (blockchain metrics, for crypto), Macro (economic context), Quant (statistical edge), and Entry Timing (execution precision).

Use the AI Committee when:

• You're considering a larger than usual position and want to pressure-test the idea from multiple angles
• The setup looks good technically but you're uncertain about the macro or sentiment backdrop
• You want an independent check before taking a trade that goes against your recent trend of losses

It's most valuable for significant decisions — not for every trade. Pay close attention to any specialist returning a Hard Block. A Hard Block from the Risk specialist is particularly important: it indicates a structural problem with the sizing or drawdown context that should override a good-looking chart.

It depends on your priorities. Here's a direct comparison:

• Use local AI when: You want complete privacy (nothing leaves your machine), you're doing private research involving your strategies, you need offline capability, or you're on a limited plan and want to reduce per-query costs
• Use remote AI when: You need the highest quality reasoning for complex multi-step analysis, you're using the AI Committee, or you're on a machine with limited RAM/CPU that makes local models impractically slow

The 1.5B local model is a good default for most traders — fast enough for session use, capable enough for most analysis, and fully private. Upgrade to remote AI for demanding tasks when you need it. You can switch between providers instantly from the AI selector in the toolbar.

No Advisory only All AI output in Dome Terminal is advisory. This is a core principle of how the product is built, not a legal disclaimer.

Risk gates, risk limits, position sizing, and all execution actions remain entirely under trader control. The AI can tell you that a setup looks technically sound, that the Risk specialist recommends smaller size, or that your last 20 journal entries show a behavioral pattern worth investigating — but it cannot decide, enter, or exit anything on your behalf without your explicit action.

The terminal is designed to make you a more informed, more disciplined trader. It is not designed to trade for you. For automated execution, bots execute strategies you define in Dome Script — and even those require you to configure the strategy, set the risk gates, review the paper logs, and manually enable live mode.

The correct sequence, in order:

• Write a Dome Script strategy in Script Library — define the entry/exit conditions, inputs, and risk parameters
• Validate it — click Validate in the Script Library editor. The script must pass all validation checks before you can create a bot from it
• Create a bot — from the validated script, click Create Bot, configure the risk gates (daily loss limit, max position, correlation gate), and set the symbol and timeframe
• Run in paper mode — paper mode is the default. Run it there for at least two calendar weeks through a variety of market conditions
• Review the paper logs critically — look for executions that don't match your intended logic, unusual loss clustering, and whether the fill behavior matches your expectations
• Consider live mode — only after reviewing paper logs thoroughly. Start with reduced position sizing and expand gradually

There are no shortcuts in this sequence. Skipping the paper period is the most common cause of unexpected losses when bots are deployed live.

Paper mode runs your bot's strategy logic against live market data — reading real prices, evaluating conditions in real time — but places simulated orders rather than real ones. It builds a complete performance record (fills, PnL, signal log) without any real money at risk.

Paper mode is realistic for evaluating strategy logic and signal timing. It is optimistic about fill quality: it assumes you get filled at or near the signal price. In live trading, fills depend on liquidity, queue position, and market conditions at execution time — which can differ meaningfully from the paper assumption, especially in fast markets.

The purpose of paper mode is to verify that the strategy fires correctly, at the right times, in the right direction — not to guarantee that live results will precisely match paper results. Treat paper performance as an upper bound on what live trading with that strategy is likely to produce.

Yes, in live mode A bot running in live mode with a connected broker account can place real orders that result in real losses. This is why risk gates exist and why the paper period is non-negotiable.

Before enabling live mode, you must configure three risk gates that act as automatic circuit breakers:

• Daily Loss Gate: The maximum loss the bot can accumulate in a single day before it pauses automatically. Set this to a number you can genuinely absorb, not an optimistic one.
• Maximum Position Gate: The maximum notional size of any single position the bot can open. Prevents runaway entries from scaling beyond your intention.
• Correlation Gate: Prevents the bot from opening positions that are highly correlated with other open positions, avoiding inadvertent concentration risk.

These gates don't eliminate risk — they bound it. Your strategy may still produce losses within those bounds. Configure the gates based on what you can genuinely afford to lose, not on what you hope won't happen.

When a risk gate triggers, the bot immediately stops executing new orders. The specific gate that fired and the reason are logged in Bot → Logs with a timestamp. No further orders are placed until you take explicit action to resume.

The bot does not automatically close existing open positions when a gate triggers. It stops opening new ones. You review the log, understand what caused the gate to fire, and manually re-enable the bot after deciding it is safe to do so.

This design is intentional: automatic position closure in response to a gate could itself cause losses (closing at the worst moment during a brief spike). The human review step before resuming is a feature, not a limitation. If a gate fires repeatedly, that's feedback that either the strategy or the gate parameters need revisiting.

The free plan gives you access to the core terminal features so you can evaluate whether Dome Terminal fits how you trade:

• Charting: Full charting with drawing tools, standard indicators, and multi-timeframe viewing
• Trade Journal: Unlimited journal entries with all standard fields and basic tagging
• Market Intelligence: Access to Market Intelligence panels (with usage limits)
• AI assistance: Limited ORACLE queries per day using the local AI model tier
• Academy: Access to all Academy learning paths and Mentor mode
• Bot paper mode: Limited bot paper mode sessions for evaluation

Features that require a paid plan include unlimited AI queries, remote AI providers, the AI Committee, unlimited bot management (including live mode), Python Lab, advanced Script Library features, and priority data provider integrations. See the Pricing page for the full feature comparison.

Yes Plan changes can be made at any time from Account → Billing.

• Upgrades: Apply immediately. You gain access to the new plan's features as soon as the upgrade is confirmed. Billing is prorated — you pay only for the remaining days of the current billing period at the new rate.
• Downgrades: Apply at the end of the current billing period. You retain full access to your current plan features until the period ends. No refund is issued for the remaining period.

If you downgrade and have active bots in live mode, those bots will pause when the downgrade takes effect if the bot management feature is not included in your new plan. You'll receive an email notification before this happens.

You can cancel anytime from Account → Billing → Cancel Subscription. No cancellation fee, no required notice period.

After cancelling:

• Your access continues until the end of the current billing period — you don't lose access immediately
• At the end of the period, your account reverts to the free plan. Your data (journal, scripts, notebooks) is retained — you won't lose anything, though some features become inaccessible until you resubscribe
• Active bots in live mode pause when the paid plan expires
• You can resubscribe at any time and pick up where you left off

If you're cancelling because something isn't working or doesn't fit your workflow, reach out to support first — we'd rather solve the problem than lose you.