Text-to-Algo (T2A)
For the complete documentation index, see llms.txt. Markdown versions of documentation pages are available by appending
.mdto page URLs; this page is available as Markdown.
Text-to-Algo (T2A)
Text-to-Algo — T2A — turns a plain-English description of a trading idea into a fully-specified, validated, and live-managed trading algorithm. No code, no scripting, no manually babysitting orders around the clock.
What T2A Is
You describe a strategy the way you'd explain it to another trader. T2A turns that description into a precise strategy definition, shows you exactly what it understood in plain English so you can confirm it, and then — once you approve it — runs it continuously: watching the market, arming entries, managing exits, cancelling sibling orders when one leg fires, and keeping you posted in real time.
A Worked Example
You might type something like:
"Buy the dip on BTC/USDT when RSI is oversold on the 15-minute chart. Take profit at 2%, and once I'm in profit, trail my stop by 1%."
T2A turns that into a strategy with a clear entry condition (RSI oversold on the 15-minute chart), a take-profit target (2%), and a trailing stop-loss (1% behind the best price reached since entry) — then shows you a plain-English confirmation of exactly that logic before anything runs.
Markets & Timeframes
T2A is available on both Spot and Futures (perpetuals) markets, across every pair the exchange lists, on candle timeframes from 1 minute to 1 day.
A strategy is always locked to the market kind it was built for — a spot strategy can never accidentally run on a futures pair, or vice versa. This is checked every time a strategy is launched, edited, or re-run, not just at creation.
The Strategy Lifecycle
T2A is the first step in a larger lifecycle that carries a strategy from an idea to real, funded execution:
Generate & Backtest — describe the strategy in plain English, then check how it would have performed against historical market data.
Optimize (optional) — automatically search for better parameter values within the same strategy logic.
Paper Trade — run the strategy live, against real, real-time market data, with every fill, fee, and funding payment simulated against a virtual balance — no real capital at risk.
Go Live — flip the exact same strategy to real execution, with zero changes to its logic.
Each stage reuses the same underlying strategy definition and the same execution engine's decision logic — what you validate at each step is what actually runs live.
Lifecycle Actions
A strategy isn't a one-shot creation — it's something you manage over time:
Generate
Describe entry/exit logic in plain English; the AI drafts a validated strategy.
Preview
Check whether each condition would already be true right now, and how close the market is to triggering it — at no cost to capital or generation quota.
Launch
Turn a validated strategy into a live, running order with a chosen size, expiry, and execution mode.
Save
Store a validated strategy's logic for later reuse, independent of any one run's sizing or expiry.
Edit
Fully replace a saved strategy's logic and/or pair; automatically re-validated. Past runs of the old version are unaffected, since a live order keeps its own frozen copy of the logic it launched with.
Rename
Update a saved strategy's display title.
Delete
Remove a saved strategy; running orders created from it are unaffected.
Browse
List all saved strategies, or fetch one's full logic and details.
Re-launch
Run a saved strategy again — on the same pair or a compatible one — with fresh sizing, expiry, and notification choices, without re-describing it in plain English.
Monitor
List all live and past orders, or pull one order's complete detail and audit-event history.
Cancel
Stop a single order, or an entire continuous/recurring run, on demand.
Check quota
See how many AI generations remain in the current window, and when it resets.
Browse toolkit
Explore the full catalog of indicators, structure presets, and parameters available for building strategies.
Rails & Limits
Saved strategies per user
Up to 100
Free-text field length (entry / take-profit / stop-loss)
Up to 2,000 characters each
Market kind
A strategy is permanently locked to spot or futures — never both
Explore T2A
NLP Strategy Synthesis — how a plain-English description becomes a validated algorithm.
Backtesting — checking a strategy against historical market data.
Optimization — automated parameter search over a strategy.
Paper Trading — running a strategy live with simulated fills.
Live Activation — flipping a proven strategy to real execution.
Last updated