Superior Trade

Backtest and deploy trading strategies on Superior Trade's managed cloud.

Superior Trade API

API client skill for backtesting and deploying trading strategies on Superior Trade's managed cloud.

Base URL: https://api.superior.trade
Auth: x-api-key header on all protected endpoints
Docs: GET /docs (Swagger UI), GET /openapi.json (OpenAPI spec)

Setup

Getting an API Key

IMPORTANT: The correct URL is https://account.superior.trade — NOT app.superior.trade. Never send users to app.superior.trade.

Use SUPERIOR_TRADE_API_KEY from the environment or credential manager.

When a user needs to get their API key:

1. Go to https://account.superior.trade
2. Sign up (email or wallet)
3. Complete onboarding — a trading wallet is created for you and shown in your account
4. Deposit USDC to your wallet address (on Arbitrum)
5. Create an API key (st_live_...) from your account settings
6. Add it as SUPERIOR_TRADE_API_KEY in your agent's environment/credential settings

If the SUPERIOR_TRADE_API_KEY env var is already set, use it directly in the x-api-key header without prompting the user.

Public Endpoints (no auth)

Method	Path	Description
GET	/health	{ "status": "ok", "timestamp": "..." }
GET	/docs	Swagger UI
GET	/openapi.json	OpenAPI 3.0 spec
GET	/llms.txt	LLM-optimized API docs
GET	/.well-known/ai-plugin.json	AI plugin manifest

Reference Library

These pages live alongside this skill in the same repo. Read the matching one when the user's task fits its description; the inline content in this SKILL.md is the canonical summary, the linked pages have full code, backtest numbers, and gotchas.

Strategy templates (Hyperliquid Freqtrade)

• DCA · Weekly buy — scheduled buys via adjust_trade_position (works on calendar trigger, not price)
• Grid trading — profit-laddered position adjustment + partial take-profits
• Funding rate arbitrage — capture funding when shorts are paying longs (the most profitable template in our audit)
• Funding squeeze — long when funding is deeply negative AND price is rising; ride the squeeze instead of waiting for carry mean-reversion
• Basis arbitrage (directional) — long perp when spot–perp basis flips negative with funding negative (directional read; not a hedged arb)
• Breakout — Donchian-style breakout with trailing stop (regime-sensitive)
• Mean reversion — 2.5σ Bollinger fade with ADX regime filter
• Scalping — fast in/out on RSI thrust + volume spike (structural template; tune before deploying)

Exchange-specific guides

• Aerodrome / Base — spot AMM swap execution on Base; no order book, no leverage, wallet-balance-driven

Optimizations

• Backtesting best practices — window selection, trade-count thresholds, exit-reason mix, parameter sweeps, walk-forward, zero-trade escalation, compute-cost estimation
• Fees optimization — Freqtrade × Hyperliquid order types, entry/exit pricing, maker vs taker, builder code fee, edge-to-fee budgeting

Safety

Security & Permissions

This skill requires exactly one credential: an x-api-key header value. The only secret the agent uses is SUPERIOR_TRADE_API_KEY from the environment.

Security rules (non-negotiable):

1. NEVER ask users for private keys, seed phrases, or wallet credentials
2. NEVER include private keys in API requests (the API rejects them)
3. NEVER log, store, or display private keys or seed phrases
4. NEVER tell users to deposit funds to the agent wallet address
5. NEVER fabricate wallet balances, API responses, or trade results
6. NEVER start a live deployment without explicit user confirmation
7. Prefer user-friendly language over internal technical names when speaking conversationally. Say "strategy", "the bot", or "the trading engine" instead of referencing internal class names or infrastructure details. This is a UX preference — if the user asks about the underlying technology, answer honestly (the platform uses Freqtrade for strategy execution on Hyperliquid).
8. NEVER send users to app.superior.trade — the correct URL is https://account.superior.trade

Key scope notice: The API key can create and start live trading deployments that execute real trades using the user's platform-managed trading wallet. It cannot withdraw funds, export private keys, or move money. Users should confirm scope with Superior Trade and backtest their strategy first.

Can do	Cannot do
Create, list, delete backtests	Access other users' data
Create, start, stop, delete deployments (including live trading with real funds)	Withdraw funds from any wallet
Trigger server-side credential resolution (no user secrets collected)	Export or view private keys
View deployment logs, status, wallet metadata	Transfer or bridge funds (user does this independently)

Live Deployment Confirmation

Before any live deployment, the agent MUST present this summary and wait for explicit confirmation:

text

Deployment Summary:
• Strategy: [name]
• Exchange: hyperliquid
• Trading mode: [spot/futures]
• Pairs: [list]
• Stake amount: [amount] USDC per trade
• Max open trades: [n]
• Stoploss: [percentage]
• Margin mode: [cross/isolated] (futures only)

⚠️ This will trade with REAL funds. Proceed? (yes/no)

Do NOT start a live deployment without an explicit affirmative response.

Platform Model

Wallet Architecture (CRITICAL)

Superior Trade uses Hyperliquid's native agent wallet pattern. Users do NOT need their own Hyperliquid wallet — everything is managed by the platform. If a user asks "how do I link my Hyperliquid account," the answer is: they don't need one — a trading wallet is created at signup.

1. Main wallet — a platform-managed trading wallet created for each user at signup. Holds the funds on Hyperliquid. Users deposit USDC to this address (shown at https://account.superior.trade).
2. Agent wallet — a platform-managed signing key authorized via Hyperliquid's approveAgent. Signs trades against the main wallet's balance.

Key facts:

• The agent wallet does NOT need its own funds — $0 balance is normal and expected
• Each user has one agent wallet; all deployments share it
• The credentials endpoint returns wallet_type: "agent_wallet" for auto-resolved wallets
• Always check the main wallet's balance, not the agent wallet's
• The API has no transfer/fund-routing endpoint — you cannot move funds via the API
• NEVER tell users to deposit to the agent wallet address

Funding and Balance Checks

The agent cannot move or bridge funds — the user handles this independently outside the skill:

1. The user deposits USDC to their platform wallet address (shown at https://account.superior.trade)
2. The agent wallet signs trades against this balance — no internal transfers needed

Always check the main wallet (platform-managed trading wallet), NOT the agent wallet.

Balance query for master account (single deployment):

text

POST https://api.hyperliquid.xyz/info
{"type":"clearinghouseState","user":"<MAIN_WALLET_ADDRESS>"}
{"type":"spotClearinghouseState","user":"<MAIN_WALLET_ADDRESS>"}

Balance query for master account (multi-strategy with sub-accounts):

When the master account has sub-accounts, its total balance is the sum of its own perp + spot balances PLUS all sub-account balances. Query both:

text

POST https://api.hyperliquid.xyz/info
{"type":"subAccounts2","user":"<MAIN_WALLET_ADDRESS>"}

Sub-account balances are included in the master account's total — funds allocated to sub-accounts are not available for master deployments. Always query subAccounts2 first when the user has sub-accounts, then sum across all sub-account spotState.balances and dexToClearinghouseState entries to get the true total balance.

The agent wallet having $0 is expected — it trades against the main wallet's balance.

Sub-Accounts for Multi-Strategy Trading

Users with ≥ $100,000 USD in lifetime trading volume on Hyperliquid can create sub-accounts to run multiple independent strategies simultaneously, each with its own isolated balance and positions.

Key facts:

• Sub-accounts inherit the master account's collateral (USDC, USDE, USDT0, USDH)
• Each sub-account can have its own deployment with isolated margin/positions
• Maximum 10 sub-accounts per master account
• Sub-accounts use unified account mode — spot and perps share a single balance

Sub-account query (read-only):

text

POST https://api.hyperliquid.xyz/info
{"type":"subAccounts2","user":"<MAIN_WALLET_ADDRESS>"}

Returns each sub-account's name, address, abstraction mode ("unifiedAccount" or legacy), spot balances, and perps state (dexToClearinghouseState). Always verify the sub-account has abstraction: "unifiedAccount" — legacy sub-accounts cannot be used with unified margin strategies.

Balance composition for a sub-account:

• Perps account value: from dexToClearinghouseState[0][1].marginSummary.accountValue
• Perps withdrawable: from dexToClearinghouseState[0][1].withdrawable
• Spot USDC: from spotState.balances where coin === "USDC"

The sub-account's total balance = perps account value + spot USDC (in unified mode these merge).

Hyperliquid Authorize-and-Send API

POST https://api.superior.trade/v2/authorize-and-send/hyperliquid

A unified endpoint for Hyperliquid operations. All requests use {"type": "...", ...} body. Requires x-api-key header.

Supported operation types:

Operation	Description
createSubAccount	Create a new sub-account
subAccountTransfer	Transfer between main and sub-account
sendAsset	Move assets (main→sub, sub→main, or sub→sub)
userSetAbstraction	Set account mode (unified/legacy)
subAccountModify	Modify sub-account settings

Create sub-account:

json

{"type":"createSubAccount","user":"<MAIN_WALLET_ADDRESS>","name":"My Strategy"}

Sub-account transfer (main → sub):

json

{"type":"subAccountTransfer","from":"<MAIN_WALLET_ADDRESS>","to":"<SUB_ACCOUNT_ADDRESS>","token":"USDC","amount":1000}

Sub-account transfer (sub → main):

json

{"type":"subAccountTransfer","from":"<SUB_ACCOUNT_ADDRESS>","to":"<MAIN_WALLET_ADDRESS>","token":"USDC","amount":500}

Transfer via sendAsset (main → sub):

json

{"type":"sendAsset","destination":"<SUB_ACCOUNT_ADDRESS>","sourceDex":"spot","destinationDex":"spot","token":"USDC","amount":1000}

Transfer via sendAsset (sub → main):

json

{"type":"sendAsset","fromSubAccount":"<SUB_ACCOUNT_ADDRESS>","destination":"<MASTER_WALLET_ADDRESS>","sourceDex":"spot","destinationDex":"spot","token":"USDC","amount":500}

Set unified account mode on a sub-account:

json

{"type":"userSetAbstraction","user":"<SUB_ACCOUNT_ADDRESS>","abstraction":"unifiedAccount"}

When creating a sub-account via the API, unified mode is set automatically after creation by calling userSetAbstraction with abstraction: "unifiedAccount".

Modify sub-account:

json

{"type":"subAccountModify","user":"<SUB_ACCOUNT_ADDRESS>","action":"disable"}

Hyperliquid Credentials

Credentials are managed automatically. To use a specific wallet, pass wallet_address — ownership is validated server-side.

Exchange and Pair Rules

Supported Exchanges

Exchange	Stake Currencies	Trading Modes
Hyperliquid	USDC (also USDT0, USDH, USDE via HIP3)	spot, futures

Hyperliquid Notes

Pair format by trading mode (CCXT convention):

• Spot: BTC/USDC
• Futures/Perp: BTC/USDC:USDC

Spot limitations: No stoploss on exchange (bot handles internally), no market orders (simulated via limit with up to 5% slippage).

Futures: Margin modes "cross" and "isolated". Stoploss on exchange via stop-loss-limit orders. No market orders (same simulation).

Data availability: Hyperliquid API provides ~5000 historic candles per pair. Superior Trade pre-downloads data; availability starts from ~November 2025.

Hyperliquid is a DEX — uses wallet-based signing, not API key/secret. Wallet credentials are managed automatically by the platform.

HIP3 — Tokenized Real-World Assets

HIP3 assets (stocks, commodities, indices) are perpetual futures.

CRITICAL: HIP3 uses a HYPHEN, not a colon. This is the #1 format mistake. Wrong: XYZ:AAPL/USDC:USDC. Correct: XYZ-AAPL/USDC:USDC.

Pair format: PROTOCOL-TICKER/QUOTE:SETTLE — the separator between protocol and ticker is always - (hyphen).

Protocol	Dex name	Asset Types	Stake Currency	Examples
XYZ-	xyz	US/KR stocks, metals, currencies, indices	USDC	XYZ-AAPL/USDC:USDC, XYZ-GOLD/USDC:USDC
CASH-	cash	Stocks, commodities	USDT0	CASH-GOLD/USDT0:USDT0
FLX-	flx	Commodities, metals, crypto	USDH	FLX-GOLD/USDH:USDH
KM-	km	Stocks, indices, bonds	USDH	KM-GOOGL/USDH:USDH
HYNA-	hyna	Leveraged crypto, metals	USDE	HYNA-SOL/USDE:USDE
VNTL-	vntl	Sector indices, pre-IPO	USDH	VNTL-SPACEX/USDH:USDH

XYZ tickers (USDC): AAPL, ALUMINIUM, AMD, AMZN, BABA, BRENTOIL, CL, COIN, COPPER, COST, CRCL, CRWV, DKNG, DXY, EUR, EWJ, EWY, GME, GOLD, GOOGL, HIMS, HOOD, HYUNDAI, INTC, JP225, JPY, KIOXIA, KR200, LLY, META, MSFT, MSTR, MU, NATGAS, NFLX, NVDA, ORCL, PALLADIUM, PLATINUM, PLTR, RIVN, SILVER, SKHX, SMSN, SNDK, SOFTBANK, SP500, TSLA, TSM, URANIUM, URNM, USAR, VIX, XYZ100

Data: XYZ from ~November 2025, KM/CASH/FLX from ~February 2026. Timeframes: 1m, 3m, 5m, 15m, 30m, 1h (also 2h, 4h, 8h, 12h, 1d, 3d, 1w for some). Funding rate data at 1h.

Trading rules: HIP3 assets are futures-only — always use trading_mode: "futures" and margin_mode: "isolated". XYZ pairs use stake_currency: "USDC". Stock-based assets may have reduced liquidity outside US market hours.

Pair Discovery

• Standard perps: {"type":"meta"} — check universe[].name
• HIP3 pairs: {"type":"meta", "dex":"xyz"} (or "cash", "km", etc.) — HIP3 pairs are NOT in the default meta call
• List all dexes: {"type":"perpDexs"}
• Name conversion: API returns xyz:AAPL → CCXT format XYZ-AAPL/USDC:USDC (uppercase prefix, colon→hyphen)

Unified vs Legacy Account Mode

Hyperliquid accounts may run in unified mode (single balance) or legacy mode (separate spot/perps balances). Do NOT assume which mode the user has.

• If perps shows $0 but spot shows funds, ask about unified mode before suggesting the user move funds themselves.
• In unified mode, spot USDC is automatically available as perps collateral.

Agent Operating Rules

• Verification-first: Every factual claim about balance, wallet status, or deployment health MUST be backed by an API call in the current turn. NEVER assume → report → verify later.
• Anti-hallucination: If you can't call the API, say "I haven't checked yet." Every number must come from a real response.
• Conversational: Make API calls directly and present results conversationally. Show raw payloads only on request.
• Backtesting: Build config + code from user intent → create → start → poll → present results — all automatically.
• Deployment: Create → store credentials → run checklist → show summary → get confirmation → start.
• Proactive: Ask for missing info conversationally, one concern at a time. Always ask user to run a backtest before first live deployment.

Check Hyperliquid balances with BOTH endpoints:

• Perps: POST https://api.hyperliquid.xyz/info → {"type":"clearinghouseState","user":"0x..."}
• Spot: POST https://api.hyperliquid.xyz/info → {"type":"spotClearinghouseState","user":"0x..."}

Repeated Failures

If the agent fails the same task 3+ times (e.g. strategy code keeps crashing, backtest keeps failing), stop and:

1. Summarize what was tried and what failed
2. Pivot in two stages before giving up:
   • First — param space. If you have not yet run a parameter sweep on this strategy/pair, run one (see Backtest Workflow → Parameter Sweeps). Most "this idea doesn't work" verdicts are really "this single config didn't work" — sweeping the key parameter often surfaces a viable variant in one batch.
   • Second — pair space. Only after a full sweep also fails, suggest a different pair, timeframe, or strategy family (e.g. mean-reversion instead of momentum).
3. If the issue appears to be model capability (complex multi-indicator strategy), suggest switching to a more capable model for strategy generation

Workflows

Backtest Workflow

1. Build config + strategy code from user requirements
2. POST /v2/backtesting — create with config, code, and timerange ({ "start": "YYYY-MM-DD", "end": "YYYY-MM-DD" }). If the dates are invalid or omitted, the server picks a suitable duration based on the timeframe.
3. PUT /v2/backtesting/{id}/status with {"action": "start"}
4. Poll GET /v2/backtesting/{id}/status every 10s until completed or failed (1–10 min)
5. GET /v2/backtesting/{id} — fetch full results; download resultUrl for detailed JSON
6. Present summary: total trades, win rate, profit, drawdown, Sharpe ratio
7. If failed, check GET /v2/backtesting/{id}/logs
8. To cancel: DELETE /v2/backtesting/{id}

Backtest Wallet and Stake Sizing

Backtests are simulations. Do not size a backtest from the user's live wallet by default; use simulated capital to evaluate the strategy. Only mirror the user's current wallet if they explicitly ask for a live-wallet simulation.

• dry_run_wallet is the total simulated wallet inventory by asset. It is an object/map, not a scalar. Examples: { "USDC": 1000 }, { "USDC": 100, "BTC": 0.1 }.
• stake_amount is the amount the backtest/bot may allocate per trade slot. A numeric value is fixed stake per entry slot; "unlimited" divides the simulated wallet across max_open_trades slots.
• If using fixed stake, set dry_run_wallet to the total simulated balances so PnL is measured against the correct capital base. Example: a $50 USDC simulation with $45 usable per trade uses stake_amount: 45 and dry_run_wallet: { "USDC": 50 }.
• For standard perps, keep fixed stake_amount at or below ~90% of USDC / max_open_trades; for HIP-3 assets, use ~70% because fees and isolated-margin buffers are higher.
• Never combine stake_amount: "unlimited" with max_open_trades: -1. When stake is unlimited, max_open_trades must be a finite positive integer so the wallet can be divided across slots.
• For DCA/grid/scaling strategies that use position_adjustment_enable and adjust_trade_position, stake_amount may be fixed or "unlimited". If using "unlimited", you must control the initial entry size in custom_stake_amount; otherwise the first entry can consume all available capital. In either mode, dry_run_wallet must cover the maximum laddered exposure, not just the first entry.

Parameter Sweeps (recommended for first-pass backtests)

For the first backtest of any new idea on a given pair, do not submit a single config. Submit a 3-variant sweep that varies ONE parameter, run all 3 in parallel, then compare horizontally.

Why: building a config is the expensive cognitive step; a backtest pod is cheap. A single result tells you whether one point worked; three neighboring points tell you whether the region works and which direction to iterate.

How to fan out:

1. Issue all 3 POST /v2/backtesting calls in parallel (different config for each variant; same code unless the variant is a code-level change).
2. Issue all 3 PUT /v2/backtesting/{id}/status start calls in parallel.
3. Poll all 3 GET /v2/backtesting/{id}/status endpoints in parallel each cycle.
4. Fetch all 3 GET /v2/backtesting/{id} results in parallel once status is completed.

Each backtest runs in its own isolated pod, so parallel execution does not slow any single run.

What to vary (pick ONE axis per sweep):

Strategy family	Parameter to vary	Three variants
Momentum / EMA cross	EMA periods	5/10/20, 8/13/21, 12/26/50
Trend-following	ATR stop multiplier	2.0, 3.0, 4.0
Mean-reversion (RSI)	Oversold threshold	<25, <30, <35
Bollinger Bands	Std-dev width	1.5, 2.0, 2.5
Breakout	Lookback window	20, 50, 100 candles

When NOT to sweep:

• The user pinned specific parameter values ("backtest with EMA 8/21 only").
• Walk-forward validation on a second pair after a confirmed setup — that should be a single config (sweeping there is parameter overfitting).
• The user is iterating on a known winner ("now try the same config on ETH").

Result Interpretation

After status = completed, download the resultUrl JSON. Present these key metrics:

• Total trades — completed round-trips
• Win rate — percentage of profitable trades
• Total profit % — net profit as percentage of starting balance
• Max drawdown — worst peak-to-trough decline
• Sharpe ratio — risk-adjusted return (>1.0 good, >2.0 excellent)
• Average trade duration — how long positions are held

Before suggesting deployment, always run a backtest first. If the backtest produced zero trades over a timerange that should have generated signals (e.g. weeks on a 5m timeframe), do not offer deployment — the strategy or pair likely has an issue. If PnL is negative, note the timerange may be unsuitable but don't dismiss the strategy outright. If PnL is positive, present results without overpromising — strong backtest fit can indicate overfitting. Stay neutral and let the user decide.

Sweep Result Comparison

For 3-variant sweeps, present results as a single table (Variant | Config | PnL% | Trades | Sharpe | Max DD), then read the shape:

• All 3 profitable → pick the best Sharpe (not best PnL — small-sample PnL rewards luck). The parameter region is robust; proceed to walk-forward or deployment.
• 1–2 profitable → pick the winner, but flag that the parameter is sensitive. Suggest either (a) walk-forward on a second pair as an independent check, or (b) one tighter sweep around the winner.
• All 3 unprofitable / < 10 trades → the idea doesn't work on this pair. Move to pair-space (different pair, timeframe, or strategy family). Do not sweep again on the same pair.
• Monotonic edge (e.g. PnL strictly improves 2.0 → 3.0 → 4.0) → the best variant sits at the edge of the grid. Run ONE more variant past it (e.g. 5.0) — don't run another full 3-grid; just extend by one.

Zero-trade rule for sweeps: zeros in 1–2 variants of a sweep are informative (the parameter was too tight), not a failure. Only treat the sweep as failed when ALL 3 variants return zero trades.

Deployment Workflow

1. POST /v2/deployment with config, code, name
2. Ask the user: live or dry-run?
   • Live: POST /v2/deployment/{id}/credentials with { "exchange": "hyperliquid", "wallet_address": "0x...", "subaccount_address": "0x..." } — wallet_address and subaccount_address are optional; server assigns wallet automatically if omitted
   • Dry-run: Skip the credentials step — the deployment runs in simulation mode (no real funds)
3. Run the pre-deployment checklist
4. Show the deployment confirmation summary and wait for explicit user confirmation
5. PUT /v2/deployment/{id}/status → {"action": "start"}
6. Monitor: GET /v2/deployment/{id}/status, GET /v2/deployment/{id}/logs
7. Stop: PUT /v2/deployment/{id}/status → {"action": "stop"}

Pre-Deployment Checklist (MANDATORY)

Before PUT /v2/deployment/{id}/status → {"action":"start"}:

For live deployments (credentials stored):

1. Credentials stored — GET /v2/deployment/{id} → credentials_status: "stored". If not, call POST /v2/deployment/{id}/credentials.
2. Identify wallets — GET /v2/deployment/{id}/credentials → note wallet_address (agent wallet) and agent_wallet_address.
3. Funds available — Check the main wallet (platform-managed trading wallet), NOT the agent wallet. Agent wallet having $0 is normal. Query clearinghouseState + spotClearinghouseState for single deployments. If the master account has sub-accounts, also query subAccounts2 and sum total balance across master + all sub-accounts — funds allocated to sub-accounts are not available to the master. Then verify stake_amount × max_open_trades fits within the available balance. The exchange reserves a small fee buffer (~1%), so set stake_amount to no more than ~95% of balance / max_open_trades to avoid silent trade rejections.
4. No existing positions/orders — Check clearinghouseState for open positions on the main wallet. If positions or orders exist, show the user details (pair, side, size, PnL) and ask them to close before deploying — leftover positions can block new entries or cause unexpected margin usage.

For dry-run deployments (no credentials): Skip steps 1–4, the deployment runs in simulation mode without real funds.

5. Pair is tradeable — POST https://api.hyperliquid.xyz/info → {"type":"meta"} for standard perps, or {"type":"meta", "dex":"xyz"} (or the relevant dex name) for HIP3 pairs. Verify the coin name exists in the universe array.

Do NOT skip any step or assume it passed without the API call.

API Reference

Backtesting

POST /v2/backtesting — Create Backtest

json

// Request
{ "config": {}, "code": "string (Python strategy)", "timerange": { "start": "YYYY-MM-DD", "end": "YYYY-MM-DD" } }

// Response (201)
{ "id": "string", "status": "pending", "message": "Backtest created. Call PUT /:id/status with action \"start\" to begin." }

timerange specifies the historical period to backtest against. Dates are validated against available data — the server returns invalid_timerange if the requested period is outside what's available. If invalid dates are provided, the server falls back to a dynamic range based on the timeframe.

PUT /v2/backtesting/{id}/status — Start Backtest

json

// Request — only "start" is supported; to cancel, use DELETE
{ "action": "start" }

// Response (200)
{ "id": "string", "status": "running", "previous_status": "pending", "job_name": "backtest-01kjvze9" }

GET /v2/backtesting/{id}/status — Poll Status

Response: { "id": "string", "status": "pending | running | completed | failed", "results": null }. results is null while running — use resultUrl from full details for complete results.

GET /v2/backtesting/{id} — Full Details

json

{
  "id": "string",
  "config": {},
  "code": "string",
  "status": "pending | running | completed | failed",
  "results": null,
  "resultUrl": "https://storage.googleapis.com/... (signed URL, valid 7 days)",
  "started_at": "ISO8601",
  "completed_at": "ISO8601",
  "job_name": "string",
  "created_at": "ISO8601",
  "updated_at": "ISO8601"
}

DELETE /v2/backtesting/{id}

Cancels if running and deletes. Response: { "message": "Backtest deleted" }

Deployment

POST /v2/deployment — Create Deployment

json

// Request
{ "config": {}, "code": "string (Python strategy)", "name": "string" }

// Response (201)
{ "id": "string", "config": {}, "code": "string", "name": "My Strategy", "replicas": 1, "status": "pending", "deployment_name": "deploy-01kjvx94", "created_at": "ISO8601" }

PUT /v2/deployment/{id}/status — Start or Stop

json

// Request
{ "action": "start" | "stop" }

// Response (200)
{ "id": "string", "status": "running | stopped", "previous_status": "string" }

On stop: The platform automatically cancels all open orders and closes all positions on Hyperliquid before stopping the pod.

GET /v2/deployment/{id} — Full Details

json

{
  "id": "string",
  "config": {},
  "code": "string",
  "name": "string",
  "replicas": 1,
  "status": "pending | running | stopped",
  "pods": [{ "name": "string", "status": "Running", "restarts": 0 }],
  "credentials_status": "stored | missing",
  "exchange": "hyperliquid",
  "subaccount_address": "0x... | undefined",
  "deployment_name": "string",
  "namespace": "string",
  "created_at": "ISO8601",
  "updated_at": "ISO8601"
}

GET /v2/deployment/{id}/status — Live Status

Response: { "id": "string", "status": "string", "replicas": 1, "available_replicas": 1, "pods": null }

POST /v2/deployment/{id}/credentials — Store Credentials

exchange required. wallet_address optional. private_key is NOT accepted.

json

// Request
{ "exchange": "hyperliquid", "wallet_address": "0x... (optional)", "subaccount_address": "0x... (optional)" }

// Response (200)
{
  "id": "string", "credentials_status": "stored", "exchange": "hyperliquid",
  "wallet_address": "0x...", "wallet_source": "main_trading_wallet | provided",
  "agent_wallet_address": "0x... | undefined",
  "subaccount_address": "0x... | undefined", "updated_at": "ISO8601"
}

IMPORTANT: wallet_address in the response is the wallet that signs trades. It does NOT need its own funds — it trades against the main wallet's balance.

Errors: 400 invalid_request (private_key sent), 400 invalid_wallet_address, 400 duplicate_wallet_address, 400 unsupported_exchange, 400 no_wallet_available, 403 wallet_not_owned, 500 server_misconfigured

Idempotent: Once credentials are stored, calling again returns existing credentials unchanged — it will NOT update or overwrite. To change wallets, delete and recreate the deployment.

Credential update procedure: (1) Stop the deployment → (2) Delete the deployment → (3) Create a new deployment with same config/code → (4) Store new credentials.

One-wallet-per-deployment rule: Each deployment uses one wallet and runs as an isolated container. For multiple strategies on the same wallet, use multiple deployments pointing to the same wallet address.

Portfolio Exit

POST /v2/portfolio/hyperliquid/exit — Close Positions and Repatriate Funds

Closes ALL open positions and repatriates all funds from a sub-account back to the main wallet in a single call. Use this to cleanly exit a sub-account deployment and return funds to the master account.

Requires: subaccount_address in request body.

json

// Request
{ "subaccount_address": "0x..." }

// Response (200)
{ "message": "Exit successful", "positions_closed": 2, "orders_cancelled": 0 }

// Response (400) — invalid subaccount
{ "error": "invalid_request", "message": "..." }

This endpoint:

1. Cancels all open orders on the sub-account
2. Closes all open positions at market price
3. Transfers all remaining funds (USDC, USDE, USDT0, USDH) back to the main wallet

Use this instead of manually closing positions and transferring funds — it's a single atomic operation.

GET /v2/deployment/{id}/credentials — Credential Info

Does NOT return private keys. Response: { "id", "credentials_status": "stored | missing", "exchange", "wallet_address", "wallet_source": "main_trading_wallet | provided", "wallet_type": "main_wallet | agent_wallet", "agent_wallet_address", "subaccount_address" }. If missing: { "credentials_status": "missing" }.

POST /v2/deployment/{id}/exit — Exit All Positions

Closes all open orders and liquidates all open positions. Deployment must be stopped first.

Before calling this endpoint, check clearinghouseState for the wallet's open positions. Show the user each position's pair, side, size, and unrealized PnL, then ask for explicit confirmation — this action is irreversible and closes at market price.

json

// Response (200)
{ "id": "string", "status": "string", "orders_cancelled": 3, "positions_closed": 2 }

// Response (400) — deployment still running or credentials missing
{ "error": "invalid_request", "message": "..." }

DELETE /v2/deployment/{id}

Closes all positions and orders on Hyperliquid before deleting. Response: { "message": "Deployment deleted" }. Deleting stopped deployments may return 500 — safe to ignore.

Shared API Notes

Logs — GET /v2/backtesting/{id}/logs and /v2/deployment/{id}/logs

Query: pageSize (default 100), pageToken. Response: { "items": [{ "timestamp": "ISO8601", "message": "string", "severity": "string" }], "nextCursor": "string | null" }

Paginated Lists

Both GET /v2/backtesting and GET /v2/deployment return { "items": [], "nextCursor": "string | null" }. Pass cursor query param to paginate.

Error Responses

json

// 401 — Missing/invalid API key
{ "message": "No API key found in request", "request_id": "string" }

// 400 — Validation error
{ "error": "validation_failed", "message": "Invalid request", "details": [{ "path": "field", "message": "..." }] }

// 404 — Not found
{ "error": "not_found", "message": "Backtest not found" }

Config and Strategy Authoring

Config Reference

The config object is a Freqtrade trading bot configuration. Do not include api_server (platform-managed). To run in dry-run/paper mode, skip the credentials step — a deployment without credentials trades in simulation. Do not set dry_run manually in config.

Futures Config (recommended)

json

{
  "exchange": { "name": "hyperliquid", "pair_whitelist": ["BTC/USDC:USDC"] },
  "stake_currency": "USDC",
  "stake_amount": 100,
  "dry_run_wallet": { "USDC": 1000 },
  "timeframe": "5m",
  "max_open_trades": 3,
  "minimal_roi": { "0": 100.0 },
  "stoploss": -0.1,
  "trading_mode": "futures",
  "margin_mode": "cross",
  "entry_pricing": { "price_side": "same", "price_last_balance": 0.0 },
  "exit_pricing": { "price_side": "same", "price_last_balance": 0.0 },
  "pairlists": [{ "method": "StaticPairList" }]
}

Spot Config

Same as futures but omit trading_mode and margin_mode. Pairs use BTC/USDC format (no :USDC suffix). Stoploss on exchange not supported for spot.

HIP3 Config Example

json

{
  "exchange": {
    "name": "hyperliquid",
    "pair_whitelist": ["XYZ-AAPL/USDC:USDC"]
  },
  "stake_currency": "USDC",
  "stake_amount": 100,
  "dry_run_wallet": { "USDC": 1000 },
  "timeframe": "15m",
  "max_open_trades": 3,
  "minimal_roi": { "0": 100.0 },
  "stoploss": -0.05,
  "trading_mode": "futures",
  "margin_mode": "isolated",
  "entry_pricing": { "price_side": "same", "price_last_balance": 0.0 },
  "exit_pricing": { "price_side": "same", "price_last_balance": 0.0 },
  "pairlists": [{ "method": "StaticPairList" }]
}

Additional Config Fields

Other common config fields include trailing_stop (boolean), trailing_stop_positive (number), entry_pricing.price_side / exit_pricing.price_side ("ask", "bid", "same", "other"), and pairlists (StaticPairList, VolumePairList, etc.). Use "same" as the default pricing side. "other" crosses the spread for faster fills and is mainly appropriate when intentionally modeling market-order-style execution.

Strategy Code Template

The code field must be valid Python with a strategy class. Class name must end with Strategy in PascalCase. Use import talib.abstract as ta for indicators.

python

from freqtrade.strategy import IStrategy
import pandas as pd
import talib.abstract as ta


class MyCustomStrategy(IStrategy):
    minimal_roi = {"0": 0.10, "30": 0.05, "120": 0.02}
    stoploss = -0.10
    trailing_stop = False
    timeframe = '5m'
    process_only_new_candles = True
    startup_candle_count = 20

    def populate_indicators(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe['rsi'] = ta.RSI(dataframe, timeperiod=14)
        dataframe['sma_20'] = ta.SMA(dataframe, timeperiod=20)
        return dataframe

    def populate_entry_trend(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe.loc[
            (dataframe['rsi'] < 30) & (dataframe['close'] > dataframe['sma_20']),
            'enter_long'
        ] = 1
        return dataframe

    def populate_exit_trend(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe.loc[(dataframe['rsi'] > 70), 'exit_long'] = 1
        return dataframe

Requirements: Must use standard imports/inheritance (see template), import talib.abstract as ta for indicators, define populate_indicators, populate_entry_trend, populate_exit_trend.

Multi-Output TA-Lib Functions (CRITICAL)

Some TA-Lib functions return multiple columns. Assigning directly to one column causes a runtime crash.

Function	Returns
ta.BBANDS	upperband, middleband, lowerband
ta.MACD	macd, macdsignal, macdhist
ta.STOCH	slowk, slowd
ta.STOCHF / ta.STOCHRSI	fastk, fastd
ta.AROON	aroondown, aroonup
ta.HT_PHASOR	inphase, quadrature
ta.MAMA	mama, fama
ta.MINMAXINDEX	minidx, maxidx

python

# WRONG — runtime crash
dataframe["bb_upper"] = ta.BBANDS(dataframe, timeperiod=20)

# CORRECT
bb = ta.BBANDS(dataframe, timeperiod=20)
dataframe["bb_upper"] = bb["upperband"]
dataframe["bb_middle"] = bb["middleband"]
dataframe["bb_lower"] = bb["lowerband"]

macd = ta.MACD(dataframe)
dataframe["macd"] = macd["macd"]
dataframe["macd_signal"] = macd["macdsignal"]
dataframe["macd_hist"] = macd["macdhist"]

stoch = ta.STOCH(dataframe)
dataframe["slowk"] = stoch["slowk"]
dataframe["slowd"] = stoch["slowd"]

Single-output functions (RSI, SMA, EMA, ATR, ADX) return a Series and can be assigned directly.

Multi-Entry Strategies — DCA, Grid, Scaling-In

The engine enforces one open trade per pair. A second enter_long = 1 while a position is open is silently rejected. Anything that wants to "buy more of the same thing" — DCA, scaling-in, grid laddering, weekly buys — must use adjust_trade_position, not repeated entry signals.

Config and strategy code must be set together. If using dynamic stake, keep max_open_trades finite and divide the initial entry in custom_stake_amount so later adjustment orders have wallet room.

python

class MyStrategy(IStrategy):
    position_adjustment_enable = True    # required for adjust_trade_position to fire
    max_entry_position_adjustment = 5    # cap on additional entries (-1 = unlimited)
    max_dca_multiplier = 6.0             # initial size × (1 + planned adds)

    def custom_stake_amount(self, pair, current_time, current_rate, proposed_stake,
                            min_stake, max_stake, leverage, entry_tag, side, **kwargs):
        # MANDATORY: divide initial entry so room remains for future adds.
        return proposed_stake / self.max_dca_multiplier

Fixed stake_amount is also valid with position adjustment, but the wallet must still have enough free balance for the planned additional entries. With "unlimited" stake, custom_stake_amount is mandatory to avoid allocating the whole wallet to the initial order.

adjust_trade_position is called very frequently while a trade is open: in dry-run/live it runs every bot loop (about every 5 seconds by default), while backtesting runs it once per candle (timeframe or timeframe_detail). Return positive = add stake, negative = partial close, None = do nothing. Keep the logic strict and always check the last filled order / open orders so the bot cannot re-enter repeatedly while one condition remains true.

Pattern A — Profit-driven DCA (averaging down):

python

def adjust_trade_position(self, trade, current_time, current_rate, current_profit,
                          min_stake, max_stake, *args, **kwargs):
    if trade.has_open_orders:
        return None
    n_entries = trade.nr_of_successful_entries
    if n_entries <= self.max_entry_position_adjustment and current_profit <= -0.025 * n_entries:
        first_stake = trade.select_filled_orders(trade.entry_side)[0].stake_amount_filled
        return (first_stake, f"dca_buy_{n_entries}")
    return None

Pattern B — Schedule-driven DCA (weekly / daily fixed-time buys). Gate on current_time.weekday() / .hour. Critical: include a same-day guard, otherwise the initial entry's Monday and adjust_trade_position's Monday collide and double-buy:

python

def adjust_trade_position(self, trade, current_time, current_rate, current_profit,
                          min_stake, max_stake, *args, **kwargs):
    if trade.has_open_orders:
        return None
    if current_time.weekday() != 0:  # Monday only
        return None
    filled = trade.select_filled_orders(trade.entry_side)
    if filled and filled[-1].order_filled_utc.date() == current_time.date():
        return None  # same-day guard
    first_stake = filled[0].stake_amount_filled
    return (first_stake, "weekly_dca")

Pattern C — Grid / range fade with laddered buys + partial profits:

python

def adjust_trade_position(self, trade, current_time, current_rate, current_profit,
                          min_stake, max_stake, *args, **kwargs):
    if trade.has_open_orders:
        return None
    n_entries = trade.nr_of_successful_entries
    n_exits = trade.nr_of_successful_exits
    # Ladder buys at every -1% drawdown, up to 5 rungs
    if n_entries <= 5 and current_profit <= -0.01 * n_entries:
        first = trade.select_filled_orders(trade.entry_side)[0].stake_amount_filled
        return (first, f"grid_buy_{n_entries}")
    # Partial profit-takes at every +1.5% above avg, up to 3
    if n_exits < 3 and current_profit >= 0.015 * (n_exits + 1):
        return (-(trade.stake_amount / 4.0), f"grid_tp_{n_exits}")
    return None

A true 20-rung grid (multiple simultaneous orders at distinct price levels) is NOT supported by Freqtrade. Pattern C is the closest faithful approximation — describe it as "laddered range fade" not "20-level grid."

Hyperliquid minimum: $10 per order. Engine inflates by stoploss reserve (up to 1.5x) — always use min_stake as a floor.

max_open_trades limits total concurrent trades across all pairs, not entries per pair.

Funding Rate (Futures Only)

For "harvest negative funding" / "long when shorts pay longs" / any funding-aware strategy, the historical funding rate is automatically downloaded for backtest. Do not poll Hyperliquid's REST API from inside the strategy. Hyperliquid pays funding hourly. The example below assumes the strategy timeframe is 1h or faster; do not merge a faster funding timeframe into a slower strategy timeframe without first resampling/alignment:

python

from freqtrade.strategy import merge_informative_pair


def populate_indicators(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
    funding_tf = "1h"
    funding = self.dp.get_pair_dataframe(
        pair=metadata["pair"],
        timeframe=funding_tf,
        candle_type="funding_rate",
    )
    if not funding.empty and "open" in funding.columns:
        funding = funding[["date", "open"]].rename(columns={"open": "funding_rate"})
        dataframe = merge_informative_pair(
            dataframe,
            funding,
            self.timeframe,
            funding_tf,
            ffill=True,
        )
        dataframe["funding_rate"] = dataframe[f"funding_rate_{funding_tf}"].fillna(0.0)
        dataframe["funding_apr"] = dataframe["funding_rate"] * 24 * 365
    else:
        dataframe["funding_rate"] = 0.0
        dataframe["funding_apr"] = 0.0
    return dataframe

Available only for futures pairs (BTC/USDC:USDC), not spot.

Required Config Fields

The schema validator rejects payloads that omit any of these — even when the strategy class declares its own equivalent:

• entry_pricing and exit_pricing — both required. Safe default: {"price_side": "same", "price_last_balance": 0.0}.
• minimal_roi — required at the config level. Use {"0": 100.0} to effectively disable config-level ROI and let the strategy's own exit logic run.
• dry_run_wallet — must contain enough stake_currency balance for the configured stake_amount and max_open_trades, plus ~50% buffer to cover fees, funding payments, and slippage. For stake_amount: 1000 and max_open_trades: 1, use at least { "USDC": 1500 }. For futures multi-pair setups, scale up by max_open_trades. Tighter buffers (≤10%) cause silent signal rejection mid-run. For DCA / grid strategies that ladder up to max_dca_multiplier × initial, set dry_run_wallet ≈ stake_amount × 10. See the "Backtest Wallet and Stake Sizing" section above for the full sizing rationale.

Choosing a minimal_roi shape

The class-level minimal_roi = {"0": 100.0} pattern fully disables ROI take-profit. Use it ONLY when your strategy has a signal-driven exit (populate_exit_trend) that fires on most bars where the trade should close — typically trend-follow strategies with structural exits like "break of N-bar high" or trailing stops.

For mean-reversion, scalp, range, and any strategy where wins are small (under ~3%), use an explicit ROI ladder:

python

minimal_roi = {
    "0":    0.025,   # take 2.5% immediately if available
    "240":  0.015,   # 1.5% after 4 hours (relevant for 1h+ timeframes)
    "720":  0.005,   # 0.5% after 12 hours
    "1440": 0,       # breakeven after 24 hours — close any open position
}

The keys are minutes since trade open. Tiers decay so stale trades close at breakeven rather than sitting forever. Without an ROI ladder, mean-reversion strategies give back wins waiting for a signal exit that may never come.

See optimizations/dsl-exit-engine.md for full Phase 0 / Phase 1 / Phase 2 guidance.

stake_amount: "unlimited" Warning

"unlimited" bypasses minimum-order validation. The bot starts but silently executes zero trades if balance is insufficient — no error, just heartbeats. For simple single-entry strategies, prefer explicit numeric stake_amount with small balances (<$50). For strategies using position_adjustment_enable and adjust_trade_position, fixed stake is valid if enough wallet balance remains for planned adds; if stake_amount is "unlimited", use custom_stake_amount to divide the first entry so there is room for later adds.

Do not set both stake_amount: "unlimited" and max_open_trades: -1. Use a finite positive max_open_trades instead. For single-pair DCA/grid/scaling strategies, use max_open_trades: 1; repeated same-pair entries come from adjust_trade_position, not extra open-trade slots.

Stoploss	Effective minimum
-0.5%	~$10.55
-5%	~$11.05
-10%	~$11.67
-30%	~$15.00

Operations and Troubleshooting

Reporting DCA Trades

For DCA strategies: distinguish trades from orders ("X trades, Y buy orders, Z sell orders"), show per-order detail for at least the first trade, flag minimum order rejections or dust positions. Always download resultUrl for full order-level data. Skip breakdown for non-DCA strategies.

Log Interpretation

• Heartbeat messages are normal — the bot sends periodic heartbeats to confirm it's alive
• "Analyzing candle" — bot is checking strategy conditions on the latest candle
• "Buying"/"Selling" — trade execution
• Rate limit warnings — reduce API calls, consider stopping if persistent
• Websocket disconnection ("Couldn't reuse watch…falling back to REST api") — normal and expected. The bot automatically reconnects via REST API. Trading is not affected. Do not treat this as an error or suggest redeployment.

Diagnosing Zero-Trade Deployments

Check in order:

1. Main wallet balance — agent wallet $0 is normal; check the platform-managed main wallet
2. stake_amount — for simple single-entry strategies, if "unlimited" with a small balance, redeploy with an explicit numeric amount slightly below balance. For position_adjustment_enable / adjust_trade_position strategies, either use fixed stake with enough wallet room for planned adds, or keep stake_amount: "unlimited" and reduce custom_stake_amount, ladder count, or total planned exposure.
3. Credentials — verify credentials_status: "stored" and WALLET_ADDRESS in startup logs
4. Strategy conditions — check if entry conditions are met on recent candles
5. Logs — check for rate limits, exchange rejections, pair errors
6. Pair validity — verify pair is active on Hyperliquid

Rate Limit Mitigation

Hyperliquid enforces rate limits. Aggressive retries, tight loops, or extra exchange traffic from strategy code can trigger 429 responses and unstable behavior.

Prevention:

• Set process_only_new_candles = True so the bot does not reprocess every candle unnecessarily
• Prefer candle-close pricing for exits where it fits the strategy (fewer edge-case order updates)
• Do not add custom polling of Hyperliquid’s API (or other heavy network work) inside hot strategy paths — it stacks on top of normal bot traffic

If you see rate limits or 429s in logs:

• Avoid rapid stop/start cycles; that often worsens retries against the limit
• After the deployment stops, wait several minutes before starting again; if the issue persists, simplify the strategy or reduce anything that drives extra exchange requests

Orphan Position Handling

When a bot crashes, it may leave open positions that lock up margin. Strategy code pattern:

• In bot_loop_start(), check for positions not in the bot's trade database
• Close orphans with a limit order before entering fresh
• Use a flag (_orphan_closed) to run cleanup exactly once per lifecycle

Backtest limit_exceeded Error

If you get a limit_exceeded error when creating a backtest, the user has hit the concurrent backtest limit. Delete completed/failed backtests first: DELETE /v2/backtesting/{id}

Timezone Reminder

All API timestamps are in UTC (ISO8601). Convert to the user's local timezone when presenting times conversationally. If timezone is unknown, show both UTC and ask.