TradingView signal bots

The pipeline, and where it silently fails

Bot	Type	Price	TradingView	Free tier	Trial	Backtest	Paper	Exchanges
Gunbot	Self-hosted	$29/mo	yes	-	-	yes	yes	20
Coinrule	Cloud platform	0.02% of volume	yes	yes	-	-	-	20
Cryptohopper	Cloud platform	$29-129/mo	yes	yes	3 days, no card	yes	yes	16
PI Pionex	Exchange-native	0.05% per trade	yes	yes	-	-	-	-
WunderTrading	Cloud platform	$0-100/mo	yes	yes	7 days, no card	-	yes	18
Altrady	Cloud platform	$30-97/mo	yes	yes	5 days, no card	yes	yes	17
3Commas	Cloud platform	$15-140/mo	yes	-	3 days, no card	yes	yes	9
HaasOnline	Cloud platform	$19.99-249.99/mo	yes	-	7 days, no card	yes	yes	24
Binance Trading Bots	Exchange-native	free	yes	yes	-	-	-	-
OKX Trading Bots	Exchange-native	free	yes	yes	-	-	-	-
TR TradeSanta	Cloud platform	$14-70/mo	yes	yes	14 days, no card	-	-	8

Live values from our database - details and caveats in each review. "-" = not documented.

Every platform in this table automates the same relay: your Pine Script logic never leaves TradingView; what crosses the internet is an alert. That makes the pipeline's weakest hop the actual product. A platform that logs every webhook with a timestamp, parses idempotently, and shows per-alert order status turns silent failure into visible failure - the difference matters more than any strategy marketplace.

TradingView's own documentation is the source of truth for the alert side: webhook payloads, plan-gated frequencies, and the bar-state rules that decide whether your indicator repainted[6]. The exchange side is documented by the venues' API rate limits and 429 semantics[7]. The platform in the middle is what you are paying for - audit it as middleware, not magic.

The platform field, honestly compared

Native Pine execution: 3Commas runs Pine Script strategies directly and prices from $15/mo annual to $105/mo Expert[1]. The established option, with the incident history (2022 key leak) that makes the key security guide non-negotiable.

Free-tier entry: WunderTrading's free tier (one bot of each type, webhooks included) is the most usable zero-commitment start[2]; Cryptohopper's Pioneer tier allows 80 positions at a 10-minute minimum interval[3] - fine for slow signals, throttled for anything active.

Volume-fee pricing: Coinrule bills 0.02% of volume instead of a subscription[5] - at $10k monthly turnover that is $2/mo, cheaper than every flat plan; at scale it crosses over. Altrady (EUR 28-90/mo) sits in the middle with a 30-day refund window[4].

All of them require exchange API keys - that is the lane's structural custody trade, and the reason the no-API-key alternatives exist for grid/DCA-shaped strategies that do not need external signals.

Pricing models, on the same axis

Pricing models, on the same axis

The plan landscape on this page spans three pricing models, and the honest comparison depends on your alert frequency. Flat subscriptions (3Commas $15-105/mo annual[1], Altrady EUR 28-90[4], Cryptohopper $29-129/mo on paid tiers[3]) predict costs but gate webhook features by plan. Free tiers (WunderTrading one bot per type[2], Cryptohopper Pioneer at 10-minute intervals[3]) carry real limits that reshape strategy design. Volume fees (Coinrule 0.02%[5]) start cheaper than everything and cross over at scale - the crossover point is computable, not a matter of taste, and the bot cost calculator draws it for your numbers.

Feature gating is where vendors differentiate quietly: backtesting, trailing TP handling, and multi-bot routing shift between plans between quarters - our reviews carry dated plan tables precisely because this page's vendors repriced repeatedly in 2025-26[1][3][4].

Payload design, the underrated craft

The webhook JSON is a contract you write and the platform interprets - most silent failures live there. Conventions that survive production: a stable action vocabulary (buy/sell/close) the platform maps explicitly; the bot or strategy identifier in every payload so routing never guesses; quantity as explicit text (not computed floats with locale commas); and a client order ID you generate for idempotency. TradingView's alert templating supports all of this[6]; the platforms differ in which fields they honor and how they treat unknowns - which is why the first payload you send to any new platform should be a deliberately malformed one, watched in the logs.

The same discipline applies to alert volume: a strategy that fires on every bar close needs plan-gated frequency headroom[6] and a platform that queues rather than drops. Both facts are testable in a trial week; neither appears on a pricing page.

Tips that prevent the classic failures

• Log both ends. TradingView alert log plus platform webhook log - reconcile daily during the trial. The gap between them is where your strategy disappears.
• Make alerts idempotent. Include an alert ID in the payload and verify the platform does not double-fill on re-delivery. Restart storms happen.
• Pin bar states in Pine. Use confirmed-bar logic so signals match the backtest[6]; a strategy that "improves" when repainting is fixed was never profitable.
• Keep stops on the exchange, not in the alert stream. A webhook out should never be the only thing between a position and disaster - the venue-side stop survives every upstream outage.
• Size for slippage between alert price and fill price. The position size calculator does the risk math; the latency reality (seconds, worse in stress) does the rest.
• Price the volume-fee model against your turnover. Coinrule at 0.02%[5] versus flat plans - crossover depends on monthly volume; the bot cost calculator draws the lines.

Warnings

• Signal bots inherit the API-key custody chain - the 3Commas 2022 leak happened to exactly this class of platform. Trade-only, IP-whitelisted, per-service keys; treat any platform without paper mode as unevaluated.
• Repainting is the default state of chart art. Pine's bar-state rules are documented[6]; a backtest built on closed bars only is the minimum honesty bar before any webhook exists.
• Marketplace strategies are marketing until proven otherwise - the FTC's fixed-return marker[8] applies to "profitable alert packs" the same as to bots. The seller's incentive is the subscription, not your fills.
• TradingView plan limits are part of your capacity budget. Alert frequency and webhook availability gate the pipeline upstream[6]; a strategy that fires more alerts than the plan allows is a strategy with a silent outage schedule.
• Platform pricing shifts often (plan reshuffles across this page's vendors in 2025-26); verify the current schedule on the provider's own pricing page before committing - our reviews carry dated plan tables.

What the numbers say

• 11 platforms in this lane execute TradingView signals; 8 carry a free tier and 6 offer a trial window of up to 14 days.
• Median entry price across the subscription lanes: ~$24.49/mo/mo - against Coinrule's 0.02%-of-volume model[5], the crossover is your monthly volume, not an opinion.
• 45% offer backtesting and 55% paper trading - a webhook pipeline that cannot be paper-tested is an untested pipeline by definition.
• The how bots work guide explains the execution loop these alerts plug into; the API key security guide covers the key chain they depend on[7].

The custody trade, stated once more

This lane cannot avoid API keys - the platform must place orders as you, which means a trade-only key copy lives on vendor infrastructure for every platform in the table. That is the structural difference from the exchange-native and wallet lanes, and it should be priced into any decision between them: the 2022 3Commas leak drained accounts through exactly this chain[7], with trade-only keys, without withdrawal permissions. The mitigations are real (IP binding as venues like OKX document it, one key per service, small balances) and they are in the key security playbook - but they reduce, not remove.

So the honest decision rule for this lane: use it when external signals are the actual edge you are automating - a Pine strategy you have validated across regimes, not a marketplace pack. If your strategy is grid/DCA-shaped and mechanical, the no-key lanes execute the same ideas cheaper and with less custody surface. Signal automation is a tool for a specific job; this page exists so the job, not the marketing, makes the choice.