TL;DR: A prop firm challenge pays you to prove you can grow a funded account without breaking its loss limits, and those limits are exactly where automated strategies tend to fail. On Trigr you can run a challenge with an agent on Propr: connect Propr's trade-only API key, pick the challenge account, replay your agent's backtest under that account's daily-loss and drawdown rules, shrink sizing until it fits, then accept the risks and deploy. Propr keeps custody of the funds and enforces the rules.
What is a prop firm challenge, and why automate one?
A proprietary trading firm (prop firm) lets traders trade the firm's capital. Most firms first run a challenge: you trade an evaluation account with a set starting balance, and you must reach a profit target without breaching risk limits such as a maximum daily loss and a maximum drawdown. Pass, and you move to a funded account that keeps the same kind of limits.
The rules reward consistency far more than big wins. One bad day can end an account that would have been profitable over a month. That is why automation is attractive: a rule-based strategy does not revenge-trade after a loss, oversize after a win, or skip its stop.
It is also why automation is dangerous. A strategy that is fine on a personal account, where a 25% drawdown is painful but survivable, can fail a challenge in its first week. The limits change what a "good" strategy is.
How does Propr work with Trigr?
Propr is a prop firm that offers challenge and funded accounts with API access. On Trigr, an agent runs on one venue: Paper, Hyperliquid, or Propr. A Propr agent has the same anatomy as any other Trigr agent:
- One venue and one account. Here, a single Propr challenge or funded account.
- Up to six strategy slots, each trading a different asset, with allocation percentages that sum to at most 100%.
- Each strategy's own RISK node for position size, leverage, stops, take-profits and exits.
The division of responsibility is the important part:
| Party | What it does |
|---|---|
| Propr | Custodies the funds, runs the challenge, and enforces its rules (daily loss, maximum drawdown, allowed markets) |
| Trigr | Evaluates your strategies' signals, places and closes trades through your Propr API key, and sizes trades within the rules where it can |
| You | Choose the strategies and sizing, accept the risks, and can pause or disconnect at any time |
Because the rules are Propr's, read Propr's published rules for your account type before you start. Details such as the target, the limits and the drawdown type come from Propr, and Trigr reads them from the account you connect rather than assuming them.
How do you connect a Propr account safely?
The connection uses a Propr API key rather than a wallet signature.
- Sign in to Propr with the account that holds your challenge and open Settings.
- Copy your API key (it starts with
pk_live_), or generate one. Propr allows one key per account and keys carry full trading access, so there is nothing to configure. - Paste the key into Trigr's Propr setup. Trigr sends it once to its execution vault and does not show it again.
What that key can and cannot do is worth knowing precisely. It can place and close trades on your Propr account. It cannot withdraw: Propr's API offers no withdrawals or payouts, and Propr custodies the funds. To cut Trigr's access you can disconnect the agent in its settings or regenerate the key in Propr, which disconnects anything using the old key.
A few practical details:
- Pick the account. If your key can see more than one challenge, you choose which one this agent trades. An account already traded by another of your agents is shown as taken until you disconnect it there.
- The starting balance becomes the agent's capital. Until Propr reports it, the account can't be selected; try again shortly.
- Disconnecting. We recommend pausing the agent first. If the account still holds positions Trigr manages, the disconnect is refused so that open exposure is never left unmanaged.
- Saved key. After a disconnect, Trigr keeps the key so your next Propr agent can skip the paste step. You can tell it to forget the saved key.
During the beta, live agents also sit behind a one-time access password, as described in the live trading docs.
Would your strategy have passed? The rules check
This is the step that makes the biggest difference. After you pick the challenge account, Trigr replays your agent's combined backtest under that account's rules before anything is created.
How the replay works
A challenge starts at a fixed balance and ends when it passes or breaches, while a backtest spans years. Measuring the whole backtest as one challenge would make late figures meaningless. So the check treats every UTC day of the backtest as the start of its own challenge, at that day's balance, and walks it bar by bar until it passes or breaches. It then reports the worst any of those simulated challenges saw:
- Worst day, as a percentage of the balance the day began with, against the daily loss limit.
- Maximum drawdown, measured the way the account defines it (a static floor set from the starting balance, or a trailing floor that follows the high-water mark), against the drawdown limit.
- Pass statistics if the account has a profit target: how many finished starts hit the target before a limit, and after how many days on average. Funded accounts have no target, so there is nothing to pass, only rules to keep.
If any simulated start crossed a limit, the verdict is that the backtest would have failed this challenge, and the screen says how many starts breached.
Optimize sizing
If the backtest breaches, Optimize sizing scales every position down and re-runs the combined backtest until it fits the rules with a 0.5% buffer. Each click runs up to three rounds. If every strategy is already at its minimum size and the backtest still crosses a limit, it tells you so. You can also continue at your own sizing, but the confirmation screen then records that you chose a sizing whose backtest breached the rules.
Smaller positions reduce drawdowns roughly in proportion, but they also slow progress toward a profit target. The pass statistics make that trade-off visible: a sizing that never breaches but needs a very long time to reach the target may still be the wrong strategy for this account.
What the check cannot see
The replay is honest about its limits, and you should be too:
- A position held across midnight is measured one day at a time in the replay, while Propr measures the day's loss from the balance before it opened. A losing position held for several days can breach live sooner than the replay shows.
- A loss and recovery inside one bar is invisible at bar resolution, and very long backtests are sampled.
- Past results do not predict future ones. A fit is evidence, not a guarantee.
How should you size a strategy for a challenge?
Challenge limits turn sizing into the main design decision. Some rules of thumb:
- Budget for the worst day, not the average. If the daily loss limit is a few percent, a single stop-loss hit at high leverage can use most of it. Check what one full stop costs at your size and leverage.
- Count concurrent positions. Six slots that can all lose on the same day behave like one large position. Correlated crypto perps often do lose together.
- Leave room. The 0.5% buffer in the rules check is a minimum. A strategy that fits by a hair on history is likely to breach live, where costs and gaps are worse.
- Prefer strategies with defined exits. Stops, ATR trails and time stops put a ceiling on how long a loser can run. Every Trigr agent requires a strategy to keep an authored way out of a position.
The position sizing guide covers how size, leverage and exits interact, and the multi-strategy agent guide covers combining slots.
Why forward-test before risking a challenge fee?
A challenge usually costs money to enter, so a failed attempt is a real loss. Running the same strategies on a paper agent first is cheap insurance. A paper agent fills at live Hyperliquid prices, deducts trading and builder fees, and shows a frontier marker where the forward record starts, so everything after it is untouched by fitting. It models no slippage or funding, so treat it as a check on behavior rather than a forecast. Paper trading agents explains what to compare and for how long.
What happens when you deploy?
The last screen is a risk acknowledgement, and deployment is refused until you accept it. It states plainly that past performance does not predict future results, that automated software can have bugs, lose connectivity or get bad data, that Propr enforces its rules and Trigr cannot guarantee the agent never breaches them, and that a breach can fail the challenge or close the account.
Once accepted, Trigr creates the agent, links the challenge account and starts it. From then on it trades its strategies' signals on that account without further prompts. You stay in control: pause the agent at any time to block new entries, disconnect it in its settings, or regenerate the key in Propr to cut access immediately.
What does it cost?
Live trading on Propr is available on every Trigr plan, including Free; plans differ in AI credits, saved strategies and paper-agent limits, as the subscriptions and credits docs and pricing page show. Challenge fees, profit splits and account terms are set by Propr, so check them there.
Prop firm challenges are demanding by design. Backtests are not guarantees, leveraged perpetual futures can lose more than you expect, and a single breach can end an account.
Next steps
Build or pick the strategies, forward-test them on paper, then follow the venue setup in the trading agents docs. For the general path from backtest to a live account, read deploying a strategy as a trading agent.