Composer Alternative for Crypto: Trigr for Hyperliquid

TL;DR: Composer stopped supporting crypto assets and indicators on January 31, 2026 and joined SoFi on June 23, 2026, so crypto users lost a no-code way to build, backtest and automate strategies in one place. Trigr covers that loop for Hyperliquid perpetual futures: a no-code node graph or AI copilot, a point-in-time backtester with explicit costs, and paper or non-custodial live agents. It is not a drop-in replacement, since Trigr trades perps rather than spot and is not a brokerage, and for US equities Composer remains the better choice.

Composer facts below come from its own product updates page and are stated as of September 2026.

What happened to Composer's crypto support?

Composer built a following as a no-code and AI strategy builder with backtesting and automated execution. Two changes this year matter for crypto traders, both stated on Composer's product updates page:

• January 31, 2026: Composer "no longer supports crypto assets or crypto indicators."
• June 23, 2026: Composer joined SoFi, describing a community of more than 175,000 investors. Earlier updates cited $1.6 billion traded in January 2026 and $250 million in total assets.

Composer continues as a custodial brokerage for US equities and ETFs. For anyone who ran crypto strategies there, the practical outcome is simple: those strategies no longer run, and crypto indicators are gone from the builder.

What do former Composer crypto users lose?

The loss is not a single feature but the combination. Composer let you express a strategy without code, see how it would have performed, and have it execute automatically, in one product. Replacing it with separate tools usually means:

• a charting or scripting tool for the logic;
• a separate backtester, often with its own data and assumptions;
• an alert service or bot to execute trades;
• and the job of keeping all three in sync.

Each handoff is a place where the tested strategy and the running strategy can drift apart. A useful Composer alternative for crypto keeps the build, the test and the execution tied to the same definition.

How does Trigr cover the build, backtest and automate loop?

Trigr is an AI strategy builder, backtester and agent platform for Hyperliquid perpetual futures. The loop looks like this.

Build without code

Strategies are node graphs: exactly one TRIGGER, optional FILTERs, a SIGNAL, a RISK block and EXECUTE. Any node can read a different asset, so "trade ETH when BTC is above its 200-day average" is a normal graph. The RISK block sets size (1–100% of capital), leverage (1–50x), direction, take-profit and stop-loss, an ATR trailing stop and a time stop. You can build by hand in Studio, describe the strategy in plain English to the AI copilot, or have ChatGPT, Claude or Codex build it over MCP. The no-code strategy builder guide walks through each block.

Inputs go beyond price indicators: funding rate, open interest, long/short ratio, liquidations, DVOL, FRED macro series, the VIX, EIA oil stocks and CFTC Commitments of Traders data are available as building blocks. Timeframes run from 5 minutes to monthly.

Backtest with stated assumptions

Trigr's backtester follows fixed conventions, described in the backtesting docs:

• point-in-time data, so every feature uses only information available at the bar's close;
• fills at the next bar's open after a signal;
• fees and the builder fee always applied, with slippage and funding as opt-ins and the first result labeled gross;
• an audit trail flagging which inputs were real and which were simulated.

A standard backtest costs 50 credits and uses full available history. If you test many variants, an ML optimization run reports the Deflated Sharpe Ratio and the Probability of Backtest Overfitting; a plain backtest shows them as not yet computed.

Automate on your own account

A backtested strategy deploys as an agent on one venue: Paper, Hyperliquid or Propr (prop-firm challenge accounts). An agent holds up to six strategy slots on different assets, with allocations summing to at most 100%, and shows a combined backtest before you start.

On Hyperliquid, the agent trades through a trade-only agent key (an API wallet) that can place and cancel orders but cannot withdraw. Your funds stay in your own Hyperliquid account and you keep your master wallet key. Hyperliquid's documentation on API wallets describes the mechanism. The pillar on deploying a strategy as a Hyperliquid agent covers the steps.

Composer vs Trigr at a glance

What changes when you move from spot to perps?

This is the most important adjustment for former Composer crypto users. Trigr trades perpetual futures, which behave differently from holding coins:

• Funding. Perp positions pay or receive funding at regular intervals. For a strategy that holds for days or weeks, funding can matter as much as fees. Switch on funding in the backtest; Trigr uses Binance's 8-hour USDT-M historical series or a flat rate you set. Our guide to slippage and funding in perp backtests shows how gross and net can diverge.
• Leverage and liquidation. Leverage is a setting from 1x to 50x. If your old strategies were unlevered, start at 1x and long-only to approximate that exposure. Higher leverage can amplify losses quickly.
• Shorting. Perps let you go short as easily as long. That opens strategies that spot accounts cannot run, and risks they do not have.
• Minimum size. Hyperliquid rejects orders under $10 notional, so Trigr raises live entries below $11 to $11. An agent with less than $11 of capital cannot open positions.

How do you rebuild a Composer-style strategy in Trigr?

There is no importer, so rebuild the logic rather than the file:

1. Write the rules in plain English: which asset, which condition, when to exit, how much capital.
2. Map each condition to a node. The entry condition becomes the TRIGGER or SIGNAL; conditions that must also hold become FILTERs; a condition on another asset becomes a cross-asset node.
3. Set the RISK block to match your old sizing, starting unlevered.
4. Spread multi-asset ideas across agent slots. Up to six strategies on different assets can share one agent.
5. Backtest gross, then net with slippage and funding on, and read the trade log.
6. Run a paper agent before committing capital.

If your Composer logic depended on features Trigr does not offer, such as custom exit graphs or instruments other than Hyperliquid perps, check before you rebuild. Trigr's graph has fixed building blocks by design.

When is Composer still the better choice?

• You trade US equities and ETFs. Composer covers them in a brokerage account; Trigr's TradFi perps are backtest and paper only in the current beta, and they are perps, not shares.
• You want a regulated US brokerage relationship, now as part of SoFi.
• You do not want leverage or funding in the picture at all. Even at 1x, a perp is not the same as owning the asset.

Trigr is the better fit when your strategies are about crypto, you want perps and alternative data such as funding and open interest, and you want the tested strategy and the running agent to be the same thing. Backtests are not guarantees; perps are leveraged and can lose more than expected.

Next steps

Review the plans on the pricing page, then rebuild one former Composer idea as a graph and backtest it gross and net before running it on paper.