profitability-calculator
SKILL.md
profitability-calculator
1. Role
Net P&L engine for the entire Onchain x CEX Strats skill suite. No strategy skill should recommend a trade without passing through this calculator first. This is the single gate that determines whether an opportunity is worth executing.
What This Skill Computes
- Gross spread / profit between trade legs
- All cost layers itemized individually
- Net profit after subtracting every cost
- Profit-to-cost ratio as a quality metric
- Minimum spread for breakeven given the cost structure
- Sensitivity analysis across variable ranges (size, slippage, gas)
What This Skill Does NOT Do
- Execute trades -- analysis only, never sends orders
- Fetch prices -- expects prices as input from
price-feed-aggregatoror caller - Check token security -- delegate to GoPlus MCP
- Discover opportunities -- delegate to strategy skills (
cex-dex-arbitrage, etc.) - Manage positions -- no state is persisted between calls
2. Language
Match the user's language. Default: Traditional Chinese (繁體中文).
Technical labels may remain in English regardless of language:
- PnL, APY, bps, USD, ETH, gas, gwei, slippage, taker, maker
Examples:
- User writes English --> respond in English
- User writes "賺唔賺" --> respond in Traditional Chinese
- User writes "能赚多少" --> respond in Simplified Chinese
3. Account Safety
Demo Default
- Always use
okx-DEMO-simulated-tradingunless the user explicitly requests live mode. - Every output header includes
[DEMO]or[LIVE]. - Every output header includes
[RECOMMENDATION ONLY -- 不會自動執行].
Switching to Live
- User explicitly says "live", "真實帳戶", "real account", or equivalent.
- Display confirmation warning (bilingual).
- Wait for user to reply "確認" or "confirm".
- Call
system_get_capabilitiesto verifyauthenticated: true. - If authenticated: switch. If not: show
AUTH_FAILED, remain in demo.
Switching to Demo
Immediate, no confirmation needed. User says "demo" or "模擬".
4. Pre-flight
This skill has minimal dependencies. It needs:
| Dependency | Purpose | Required? |
|---|---|---|
okx-trade-mcp |
Fee schedule lookup, orderbook depth for slippage estimation | Yes |
OnchainOS CLI |
Gas price fetch (onchain-gateway gas), gas limit estimation |
Only for onchain legs |
| Fee tables (inlined below) | Static fee tables, withdrawal fee lookup | Yes (built-in) |
| Formulas (inlined below) | Calculation formulas | Yes (built-in) |
Pre-flight Checks
1. Call system_get_capabilities
- Verify authenticated: true
- Note mode: "demo" or "live"
2. If trade has onchain legs:
- Verify OnchainOS CLI is available: `which onchainos`
- If unavailable: use benchmark gas values from fee tables below
and flag confidence as "medium"
5. Skill Routing Matrix
| User Need | THIS Skill Handles | Delegate To |
|---|---|---|
| "Is this arb profitable?" | Estimate with all legs and costs | -- |
| "What are the fees on OKX?" | Full fee breakdown per leg | -- |
| "Minimum spread needed?" | Compute min-spread for breakeven | -- |
| "How does profit change with size?" | Sensitivity analysis | -- |
| "What's the current price?" | -- | price-feed-aggregator |
| "Find arb opportunities" | -- | cex-dex-arbitrage |
| "Is this token safe?" | -- | GoPlus MCP directly |
| "What's the funding rate?" | -- | funding-rate-arbitrage |
| "Execute this trade" | -- | Not supported (analysis only) |
6. Command Index
| Command | Purpose | Typical Caller |
|---|---|---|
estimate |
Compute net P&L for a set of trade legs | Strategy skills, user |
breakdown |
Detailed per-layer cost table | User (interactive) |
min-spread |
Minimum spread (bps) for breakeven between two venues | Strategy skills |
sensitivity |
How net profit varies across a parameter range | User (analysis) |
7. Parameter Reference
7.1 Command: estimate
Compute net profit/loss for a proposed multi-leg trade.
Usage
profitability-calculator estimate --legs '[
{"venue":"okx-cex","asset":"ETH","side":"buy","size_usd":1000,"order_type":"market"},
{"venue":"okx-dex","chain":"ethereum","asset":"ETH","side":"sell","size_usd":1000,"order_type":"market"}
]'
Input Schema: TradeLeg
TradeLeg:
venue: string # REQUIRED. "okx-cex" | "okx-dex" | "uniswap" | "jupiter"
chain: string # REQUIRED for DEX legs. "ethereum" | "solana" | "base" | "arbitrum" etc.
asset: string # REQUIRED. Symbol (e.g. "ETH") or contract address
side: string # REQUIRED. "buy" | "sell"
size_usd: number # REQUIRED. Trade size in USD
order_type: string # Optional. "market" (default) | "limit"
estimated_slippage_bps: number # Optional. Default: auto-calculated from orderbook
gas_gwei: number # Optional. For onchain legs. Default: fetched from onchain-gateway
withdrawal_required: boolean # Optional. Default: false
deposit_required: boolean # Optional. Default: false
bridge_required: boolean # Optional. Default: false
bridge_chain: string # Required if bridge_required=true. Target chain.
vip_tier: string # Optional. "VIP0" (default) | "VIP1" ... "VIP5"
instrument_type: string # Optional. "spot" (default) | "swap" | "futures"
funding_intervals: number # Optional. For perp legs. Number of 8h funding intervals held.
funding_rate: number # Optional. Current funding rate per 8h (e.g. 0.00015)
borrow_rate_annual: number # Optional. Annualized borrow rate for margin/short legs.
hold_days: number # Optional. Expected holding period in days. Default: 0 (instant).
Parameter Validation
| Field | Type | Validation | On Failure |
|---|---|---|---|
venue |
string | Must be one of: okx-cex, okx-dex, uniswap, jupiter, curve, pancakeswap |
INVALID_LEG |
chain |
string | Required when venue is not okx-cex. Must match Supported Chains Table below. |
MISSING_FIELD |
asset |
string | Non-empty. If contract address, must be lowercase for EVM chains. | MISSING_FIELD |
side |
string | Must be buy or sell |
INVALID_LEG |
size_usd |
number | Must be > 0 | INVALID_LEG |
order_type |
string | Must be market or limit. Default: market |
INVALID_LEG |
estimated_slippage_bps |
number | Must be >= 0. If omitted, auto-calculated. | -- |
gas_gwei |
number | Must be >= 0. If omitted, fetched live or from benchmark. | -- |
bridge_chain |
string | Required if bridge_required=true |
MISSING_FIELD |
vip_tier |
string | Must be VIP0..VIP5. Default: VIP0 |
INVALID_LEG |
instrument_type |
string | Must be spot, swap, or futures. Default: spot |
INVALID_LEG |
funding_intervals |
number | Must be >= 0. Only applies when instrument_type=swap |
-- |
funding_rate |
number | If omitted and funding_intervals > 0, fetched live via market_get_funding_rate |
FEE_LOOKUP_FAILED |
borrow_rate_annual |
number | Must be >= 0. Only applies to margin/short legs. | -- |
hold_days |
number | Must be >= 0. Default: 0 (instant execution). | -- |
Return Schema: ProfitabilityResult
ProfitabilityResult:
net_profit_usd: number # Positive = profitable, negative = loss
gross_spread_usd: number # Before any costs
total_costs_usd: number # Sum of all cost layers
cost_breakdown:
cex_trading_fee: number # size_usd * fee_rate (per CEX leg)
dex_trading_fee: number # From swap quote or estimated from protocol fee table
gas_cost: number # For onchain legs: gas_gwei * gas_limit * native_price / 1e9
slippage_cost: number # Estimated from orderbook depth or user-provided bps
withdrawal_fee: number # Fixed per-asset from OKX schedule
deposit_fee: number # Usually 0 (user pays network gas only)
bridge_fee: number # If cross-chain transfer required
funding_payment: number # For perp legs held over funding intervals
borrow_cost: number # For margin/short legs: size * borrow_rate * hold_days / 365
profit_to_cost_ratio: number # net_profit / total_costs (undefined if total_costs = 0)
is_profitable: boolean # true if net_profit_usd > 0
min_spread_for_breakeven_bps: number # Minimum spread needed to cover all costs
confidence: string # "high" | "medium" | "low"
confidence_factors:
- factor: string # Description of what affects confidence
impact: string # "reduces_confidence" | "increases_confidence"
warnings: string[] # Array of warning messages
Confidence Levels
| Level | Criteria |
|---|---|
high |
All cost data fetched live (gas, orderbook, fees). Slippage auto-calculated. |
medium |
Some data from benchmarks (e.g. gas from fee tables, not live). |
low |
User-provided estimates for multiple fields. No live data verification. |
7.2 Command: breakdown
Same input as estimate, but outputs a detailed per-layer table with additional context
per cost item (formula used, data source, alternative options).
Usage
profitability-calculator breakdown --legs '[...]'
Additional Output Fields (beyond ProfitabilityResult)
Each cost line includes:
cost_detail:
label: string # Human-readable label (e.g. "CEX Taker Fee (VIP0)")
amount_usd: number # Cost in USD
amount_bps: number # Cost in basis points relative to trade size
pct_of_gross: number # This cost as % of gross spread
formula: string # The formula used (e.g. "1000 * 0.0008 = $0.80")
data_source: string # Where the input came from ("live orderbook", "fee tables", "user input")
optimization_hint: string # Optional suggestion (e.g. "Use limit order for maker fee: 0.06%")
7.3 Command: min-spread
Calculate the minimum spread (in bps) required between two venues for a trade to be profitable after all costs.
Usage
profitability-calculator min-spread \
--venue-a okx-cex \
--venue-b okx-dex \
--chain ethereum \
--asset ETH \
--size-usd 10000 \
--vip-tier VIP0 \
--order-type market
Parameters
| Param | Required | Default | Type | Description |
|---|---|---|---|---|
--venue-a |
Yes | -- | string | First venue (buy side) |
--venue-b |
Yes | -- | string | Second venue (sell side) |
--chain |
Conditional | -- | string | Required if either venue is DEX |
--asset |
Yes | -- | string | Asset symbol or contract address |
--size-usd |
Yes | -- | number | Trade size in USD |
--vip-tier |
No | VIP0 |
string | OKX VIP tier |
--order-type |
No | market |
string | market or limit |
--include-withdrawal |
No | true |
boolean | Include withdrawal fee in cost |
--include-bridge |
No | false |
boolean | Include bridge fee in cost |
--bridge-chain |
Conditional | -- | string | Required if include-bridge=true |
Return Schema
MinSpreadResult:
min_spread_bps: number # Minimum spread needed (breakeven point)
cost_components_bps:
cex_fee_bps: number
dex_gas_bps: number
dex_fee_bps: number
slippage_bps: number
withdrawal_bps: number
bridge_bps: number
total_cost_bps: number
interpretation: string # e.g. "Need at least 18 bps spread to break even on $10,000"
7.4 Command: sensitivity
Show how net profit changes across a range of values for a given variable.
Usage
profitability-calculator sensitivity \
--legs '[...]' \
--variable size_usd \
--range '[1000, 50000]' \
--steps 10
Parameters
| Param | Required | Default | Type | Description |
|---|---|---|---|---|
--legs |
Yes | -- | TradeLeg[] | Base trade configuration |
--variable |
Yes | -- | string | Variable to sweep: size_usd, slippage, gas_gwei, vip_tier, spread_bps |
--range |
Yes | -- | [min, max] | Range of values to test |
--steps |
No | 10 |
number | Number of data points in the range |
Return Schema
SensitivityResult:
variable: string
data_points:
- value: number # Variable value at this point
net_profit_usd: number
total_costs_usd: number
profit_to_cost_ratio: number
is_profitable: boolean
breakeven_value: number # Value of variable where net_profit = 0
optimal_value: number # Value with highest profit_to_cost_ratio
summary: string # Human-readable interpretation
8. Operation Flow
Step 1: Parse and Validate
INPUT: TradeLeg[]
|
├─ Validate all REQUIRED fields present
├─ Validate field values against allowed enums
├─ Validate cross-field dependencies:
│ - DEX venue requires `chain`
│ - bridge_required=true requires `bridge_chain`
│ - funding_intervals > 0 requires instrument_type="swap"
|
├─ On validation failure → return INVALID_LEG or MISSING_FIELD
└─ On success → proceed to Step 2
Step 2: Fetch Live Cost Data
For each leg, gather the cost inputs:
FOR EACH leg IN TradeLeg[]:
IF leg.venue == "okx-cex":
├─ Fee rate: lookup from OKX Fee Tables below using vip_tier + instrument_type
│ - Spot VIP0 taker: 0.080%, maker: 0.060%
│ - Swap VIP0 taker: 0.050%, maker: 0.020%
│
├─ Slippage: if not provided, fetch orderbook:
│ market_get_orderbook(instId={ASSET}-USDT, sz="400")
│ Walk bids/asks to compute volume-weighted avg price at size_usd
│ price_impact_bps = (vwap - best_price) / best_price * 10000
│
└─ Withdrawal fee: if withdrawal_required=true
Lookup from OKX Withdrawal Fee Tables below → fixed amount per asset+network
IF leg.venue is DEX (okx-dex, uniswap, jupiter, etc.):
├─ Gas price: if gas_gwei not provided:
│ onchainos onchain-gateway gas --chain {chain}
│ On failure → use benchmark from Gas Benchmarks table below, set confidence="medium"
│
├─ Gas limit: use benchmarks from Gas Benchmarks table below:
│ - Ethereum: 150,000-300,000
│ - Arbitrum: 1,000,000-2,000,000
│ - Solana: N/A (use lamport-based calculation)
│ Or fetch via onchainos onchain-gateway gas-limit if calldata available
│
├─ DEX protocol fee: lookup from DEX Protocol Fee Tables below:
│ - Uniswap standard: 30 bps
│ - Jupiter: 0 bps (aggregator layer)
│ - Note: DEX quotes from aggregators already include protocol fees
│ so set dex_trading_fee = 0 when using aggregator quotes
│
└─ Slippage: if not provided:
Use price impact from dex-swap quote if available
Otherwise estimate from benchmarks
IF leg has funding_intervals > 0:
├─ If funding_rate not provided:
│ market_get_funding_rate(instId={ASSET}-USDT-SWAP)
│ Use fundingRate field
│
└─ funding_payment = size_usd * funding_rate * funding_intervals
(positive if receiving, negative if paying)
IF leg has borrow_rate_annual > 0 AND hold_days > 0:
└─ borrow_cost = size_usd * borrow_rate_annual * hold_days / 365
Step 3: Compute Each Cost Layer and Net Profit
# Formulas (all inlined below in Section 9)
cex_trading_fee = size_usd * fee_rate
- fee_rate: 0.0008 (taker VIP0 spot) or 0.0006 (maker VIP0 spot)
- For swaps: 0.0005 (taker VIP0) or 0.0002 (maker VIP0)
gas_cost = gas_gwei * gas_limit * native_token_price_usd / 1e9
- Example: 30 gwei * 150,000 * $3,450 / 1e9 = $15.53
- Solana: (base_fee_lamports + priority_fee_lamports) * sol_price / 1e9
slippage_cost = size_usd * estimated_slippage_bps / 10000
- Auto-calc: walk orderbook to find VWAP at size_usd
- price_impact_bps = (vwap - best_price) / best_price * 10000
withdrawal_fee = lookup from OKX Withdrawal Fee Tables below
- ETH (ERC-20): 0.00035 ETH
- ETH (Arbitrum): 0.0001 ETH
- USDT (TRC-20): 1.0 USDT
- USDT (Arbitrum): 0.1 USDT
bridge_fee = estimated from Bridge Fee Tables below
- Across Protocol: 0.04-0.12% of amount
- OKX Bridge: 0-0.1%
funding_payment = size_usd * funding_rate * funding_intervals
- Positive funding_rate + short position = receive payment
- Positive funding_rate + long position = pay funding
borrow_cost = size_usd * borrow_rate_annual * hold_days / 365
# Aggregation
total_costs = SUM(cex_trading_fee, dex_trading_fee, gas_cost,
slippage_cost, withdrawal_fee, deposit_fee,
bridge_fee, funding_payment, borrow_cost)
gross_spread = (sell_price - buy_price) / buy_price * size_usd
- Or: provided by caller as price differential
net_profit = gross_spread - total_costs
profit_to_cost_ratio = net_profit / total_costs
min_spread_for_breakeven_bps = total_costs / size_usd * 10000
Step 4: Format Output and Flag Results
IF net_profit <= 0:
→ Label: [NOT PROFITABLE]
→ Show full cost breakdown so user understands why
→ Suggest: reduce size, switch chain (lower gas), use limit orders, wait for wider spread
IF net_profit > 0 AND profit_to_cost_ratio < 2.0:
→ Label: [MARGINAL]
→ Warning: "利潤率偏低，執行風險大 / Thin margin, high execution risk"
IF net_profit > 0 AND profit_to_cost_ratio >= 2.0:
→ Label: [PROFITABLE]
ALWAYS:
→ Show cost breakdown table
→ Show confidence level and factors
→ Show next steps / related commands
→ Include disclaimer (bilingual)
9. Cost Computation Formulas
CEX-DEX Spread (basis points)
spread_bps = abs(cex_price - dex_price) / min(cex_price, dex_price) * 10000
| Variable | Definition |
|---|---|
cex_price |
Mid-price on CEX orderbook (OKX), or best bid/ask depending on direction |
dex_price |
Quote price returned by DEX aggregator for the given size |
spread_bps |
Unsigned spread in basis points (1 bp = 0.01%) |
Worked Example:
cex_price = 3,412.50 (ETH-USDT mid on OKX)
dex_price = 3,419.80 (Jupiter quote, selling ETH for USDT on Solana)
spread_bps = abs(3412.50 - 3419.80) / min(3412.50, 3419.80) * 10000
= 7.30 / 3412.50 * 10000
= 21.4 bps
Caveats:
- Always use the same quote direction (both buy or both sell) for a fair comparison.
- DEX price already includes DEX protocol fee and price impact for the quoted size.
- CEX price should use bid for sell, ask for buy -- not mid -- when evaluating executable spread.
Effective Price After Slippage
effective_price = quoted_price * (1 + slippage_bps / 10000) # for buys
effective_price = quoted_price * (1 - slippage_bps / 10000) # for sells
Price Impact Estimation from Orderbook Depth
price_impact_bps = (execution_avg_price - best_price) / best_price * 10000
Where execution_avg_price is the volume-weighted average price across consumed levels:
execution_avg_price = sum(level_price_i * level_qty_i) / sum(level_qty_i)
for levels consumed until order_size is filled
Worked Example (Selling 10 ETH):
Orderbook bids:
Level 1: 3,412.50 x 3 ETH
Level 2: 3,412.00 x 4 ETH
Level 3: 3,411.20 x 5 ETH
execution_avg_price = (3412.50*3 + 3412.00*4 + 3411.20*3) / (3+4+3)
= (10237.50 + 13648.00 + 10233.60) / 10
= 34119.10 / 10
= 3,411.91
price_impact_bps = (3412.50 - 3411.91) / 3412.50 * 10000
= 0.59 / 3412.50 * 10000
= 1.7 bps
CEX Trading Fee
cex_fee = size_usd * fee_rate
| Variable | Definition |
|---|---|
size_usd |
Notional trade size in USD |
fee_rate |
Maker or taker rate per VIP tier (see fee tables below) |
Worked Example (VIP1 Taker, $50,000 trade):
cex_fee = 50000 * 0.0007
= $35.00
DEX Gas Cost (EVM)
gas_cost_usd = gas_price_gwei * gas_limit * native_token_price_usd / 1e9
| Variable | Definition |
|---|---|
gas_price_gwei |
Current gas price in gwei (for EVM chains) |
gas_limit |
Gas units required for the transaction |
native_token_price_usd |
Current USD price of the chain's native token (ETH, BNB, MATIC, etc.) |
Worked Example (Uniswap swap on Ethereum):
gas_price_gwei = 30
gas_limit = 150,000
native_token_price_usd = 3,400
gas_cost_usd = 30 * 150000 * 3400 / 1e9
= 15,300,000,000 / 1,000,000,000
= $15.30
DEX Gas Cost (Solana)
Solana uses lamports instead of gwei:
gas_cost_usd = (base_fee_lamports + priority_fee_lamports) * sol_price_usd / 1e9
Worked Example:
tx_fee_lamports = 5,000 (base fee)
priority_fee_lamports = 50,000 (typical priority fee)
sol_price_usd = 145
gas_cost_usd = (5000 + 50000) * 145 / 1e9
= 7,975,000 / 1,000,000,000
= $0.008
Gas Caveats:
- Gas prices are highly volatile, especially on Ethereum mainnet.
- Complex swaps (multi-hop, multi-pool) consume more gas.
- L2s (Arbitrum, Base) also pay an L1 data posting fee that fluctuates.
- Always fetch real-time gas estimates; benchmarks are rough guides only.
Slippage Cost
slippage_cost_usd = size_usd * estimated_slippage_bps / 10000
Worked Example:
size_usd = 50,000
estimated_slippage_bps = 8
slippage_cost_usd = 50000 * 8 / 10000
= $40.00
Auto-calculation from orderbook:
1. Fetch orderbook: market_get_orderbook(instId, sz="400")
2. Walk levels until cumulative size >= order_size
3. vwap = sum(price_i * qty_i) / sum(qty_i)
4. price_impact_bps = abs(vwap - best_price) / best_price * 10000
5. slippage_cost = size_usd * price_impact_bps / 10000
Total Cost Aggregation
total_cost = cex_fee + dex_gas_cost + dex_protocol_fee + slippage_cost + bridge_fee + withdrawal_fee
For a round-trip trade:
total_cost_roundtrip = (cex_fee_entry + cex_fee_exit)
+ (dex_gas_entry + dex_gas_exit)
+ (dex_protocol_fee_entry + dex_protocol_fee_exit)
+ (slippage_entry + slippage_exit)
+ bridge_fee (if cross-chain)
+ withdrawal_fee (if moving between CEX/DEX)
Worked Example (CEX-DEX arb, one direction):
cex_fee (VIP1 taker) = $35.00
dex_gas (Arbitrum) = $0.30
dex_protocol_fee = $0.00 (included in quote)
slippage (CEX) = $15.00
withdrawal_fee (USDT) = $1.00
bridge_fee = $0.00 (same chain)
total_cost = 35.00 + 0.30 + 0.00 + 15.00 + 1.00 + 0.00
= $51.30
Net Profit
net_profit = gross_spread_usd - total_cost
gross_spread_usd = size_usd * spread_bps / 10000
Worked Example:
size_usd = 50,000
spread_bps = 21.4
gross_spread_usd = 50000 * 21.4 / 10000 = $107.00
total_cost = $51.30
net_profit = 107.00 - 51.30 = $55.70
Profit-to-Cost Ratio
profit_to_cost = net_profit / total_cost
Interpretation:
- < 1.0x: Costs exceed profit -- do not trade
- 1.0x - 1.5x: Marginal -- high execution risk, consider skipping
- 1.5x - 3.0x: Acceptable -- proceed with caution
- 3.0x+: Attractive -- high confidence trade
Minimum Spread for Breakeven
min_spread_bps = total_costs_usd / size_usd * 10000
Funding Payment
funding_payment = size_usd * funding_rate * funding_intervals
- Positive funding_rate + short position = receive payment
- Positive funding_rate + long position = pay funding
Borrow Cost
borrow_cost = size_usd * borrow_rate_annual * hold_days / 365
10. OKX CEX Trading Fee Tables
Spot Trading Fees
| Tier | 30d Volume (USD) | Maker | Taker |
|---|---|---|---|
| VIP0 | < 5M | 0.060% | 0.080% |
| VIP1 | >= 5M | 0.040% | 0.070% |
| VIP2 | >= 10M | 0.030% | 0.060% |
| VIP3 | >= 20M | 0.020% | 0.050% |
| VIP4 | >= 100M | 0.015% | 0.040% |
| VIP5 | >= 200M | 0.010% | 0.035% |
USDT-Margined Swap (Perpetual) Fees
| Tier | 30d Volume (USD) | Maker | Taker |
|---|---|---|---|
| VIP0 | < 5M | 0.020% | 0.050% |
| VIP1 | >= 5M | 0.015% | 0.045% |
| VIP2 | >= 10M | 0.010% | 0.040% |
| VIP3 | >= 20M | 0.008% | 0.035% |
| VIP4 | >= 100M | 0.005% | 0.030% |
| VIP5 | >= 200M | 0.002% | 0.025% |
Coin-Margined Swap / Futures Fees
| Tier | 30d Volume (USD) | Maker | Taker |
|---|---|---|---|
| VIP0 | < 5M | 0.020% | 0.050% |
| VIP1 | >= 5M | 0.015% | 0.045% |
| VIP2 | >= 10M | 0.010% | 0.040% |
| VIP3 | >= 20M | 0.008% | 0.035% |
| VIP4 | >= 100M | 0.005% | 0.030% |
| VIP5 | >= 200M | 0.002% | 0.025% |
Fee Calculation Notes
- Maker = limit order that adds liquidity to the orderbook (not immediately matched)
- Taker = market order or limit order that immediately matches
- For cost estimation, assume taker fees unless the skill specifically uses limit orders
- Fees are deducted from the received asset (spot) or from margin (derivatives)
Fee Formula
fee_usd = notional_size_usd * fee_rate
# Round-trip cost (open + close)
roundtrip_fee_usd = notional_size_usd * (entry_fee_rate + exit_fee_rate)
roundtrip_fee_bps = (entry_fee_rate + exit_fee_rate) * 10000
Quick Reference: Round-Trip Taker Fees (bps):
| Tier | Spot RT | Swap RT |
|---|---|---|
| VIP0 | 16.0 bps | 10.0 bps |
| VIP1 | 14.0 bps | 9.0 bps |
| VIP2 | 12.0 bps | 8.0 bps |
| VIP3 | 10.0 bps | 7.0 bps |
| VIP4 | 8.0 bps | 6.0 bps |
| VIP5 | 7.0 bps | 5.0 bps |
11. OKX Withdrawal Fee Tables
Withdrawal fees are flat (not percentage-based) and vary by asset and network.
BTC Withdrawals
| Network | Fee | Min Withdrawal |
|---|---|---|
| Bitcoin (BTC) | 0.0001 BTC | 0.001 BTC |
| Lightning Network | 0.000001 BTC | 0.000001 BTC |
ETH Withdrawals
| Network | Fee | Min Withdrawal |
|---|---|---|
| Ethereum (ERC-20) | 0.00035 ETH | 0.001 ETH |
| Arbitrum One | 0.0001 ETH | 0.0001 ETH |
| Optimism | 0.00004 ETH | 0.0001 ETH |
| Base | 0.00004 ETH | 0.0001 ETH |
| zkSync Era | 0.000065 ETH | 0.0001 ETH |
| Linea | 0.00015 ETH | 0.0001 ETH |
USDT Withdrawals
| Network | Fee | Min Withdrawal |
|---|---|---|
| Ethereum (ERC-20) | 3.0 USDT | 10.0 USDT |
| Tron (TRC-20) | 1.0 USDT | 0.1 USDT |
| Polygon | 0.8 USDT | 0.1 USDT |
| Arbitrum One | 0.1 USDT | 0.1 USDT |
| Optimism | 0.1 USDT | 0.1 USDT |
| Base | 0.1 USDT | 0.1 USDT |
| Solana | 1.0 USDT | 1.0 USDT |
| BSC (BEP-20) | 0.3 USDT | 10.0 USDT |
| Avalanche C-Chain | 1.0 USDT | 1.0 USDT |
| TON | 0.5 USDT | 0.1 USDT |
USDC Withdrawals
| Network | Fee | Min Withdrawal |
|---|---|---|
| Ethereum (ERC-20) | 3.0 USDC | 10.0 USDC |
| Polygon | 0.8 USDC | 0.1 USDC |
| Arbitrum One | 0.1 USDC | 0.1 USDC |
| Optimism | 0.1 USDC | 0.1 USDC |
| Base | 0.1 USDC | 0.1 USDC |
| Solana | 1.0 USDC | 1.0 USDC |
| BSC (BEP-20) | 0.3 USDC | 10.0 USDC |
| Avalanche C-Chain | 1.0 USDC | 1.0 USDC |
SOL Withdrawals
| Network | Fee | Min Withdrawal |
|---|---|---|
| Solana | 0.008 SOL | 0.1 SOL |
Other Major Assets
| Asset | Network | Fee | Min Withdrawal |
|---|---|---|---|
| BNB | BSC (BEP-20) | 0.0005 BNB | 0.01 BNB |
| MATIC | Polygon | 0.1 MATIC | 0.1 MATIC |
| AVAX | Avalanche C-Chain | 0.01 AVAX | 0.1 AVAX |
| ARB | Arbitrum One | 0.1 ARB | 1.0 ARB |
| OP | Optimism | 0.1 OP | 1.0 OP |
| OKB | X Layer | 0.001 OKB | 0.01 OKB |
Withdrawal Fee Notes
- Withdrawal fees are flat amounts, not percentages.
- Fees are updated periodically based on network conditions.
- Always verify current fees via OKX API before calculating costs.
- Internal transfers between OKX accounts (sub-accounts) are free.
- Deposits to OKX are free (user only pays network gas).
12. Gas Benchmarks per Chain
Approximate gas costs for a typical DEX swap transaction.
| Chain | Avg Gas Price | Swap Gas Limit | Approx Cost (USD) | Native Token |
|---|---|---|---|---|
| Ethereum | 20-50 gwei | 150,000-300,000 | $5.00-$30.00 | ETH |
| Arbitrum | 0.1-0.5 gwei | 1,000,000-2,000,000 | $0.10-$0.50 | ETH |
| Base | 0.005-0.02 gwei | 150,000-300,000 | $0.01-$0.05 | ETH |
| Optimism | 0.005-0.02 gwei | 150,000-300,000 | $0.01-$0.05 | ETH |
| Polygon | 30-100 gwei | 150,000-300,000 | $0.01-$0.05 | MATIC |
| BSC | 3-5 gwei | 150,000-300,000 | $0.10-$0.30 | BNB |
| Avalanche | 25-30 nAVAX | 150,000-300,000 | $0.02-$0.10 | AVAX |
| Solana | N/A (lamports) | N/A (compute units) | $0.001-$0.01 | SOL |
| X Layer | 0.01-0.05 gwei | 150,000-300,000 | $0.001-$0.01 | OKB |
Gas Notes
- Ethereum L1 gas is by far the most expensive; avoid for small trades.
- L2 costs include a small L1 data posting fee that can spike during L1 congestion.
- Multi-hop swaps (going through multiple pools) use significantly more gas.
- Approval transactions (first-time token approvals) add ~46,000 gas on EVM chains.
- Gas prices fluctuate significantly; always fetch real-time estimates.
Cost Efficiency by Chain (for $10,000 trade)
| Chain | Gas as % of Trade | Practical Minimum Trade |
|---|---|---|
| Ethereum | 0.05-0.30% | $5,000+ |
| Arbitrum | 0.001-0.005% | $100+ |
| Base | 0.0001-0.0005% | $50+ |
| Solana | 0.00001-0.0001% | $10+ |
| BSC | 0.001-0.003% | $100+ |
| Polygon | 0.0001-0.0005% | $50+ |
Solana Gas Details
- Base fee: 5,000 lamports (0.000005 SOL)
- Priority fee: 10,000-500,000 lamports depending on congestion
- Compute budget: 200,000-1,400,000 compute units
Gas Price Fallback Chain
1. Try: onchainos onchain-gateway gas --chain {chain}
2. If fails: use benchmark midpoint from table above
- Ethereum: 35 gwei
- Arbitrum: 0.3 gwei
- Base: 0.01 gwei
- Solana: 55,000 lamports
3. Flag confidence as "medium" when using fallback
13. DEX Protocol Fee Tables
Fees charged by the DEX protocol itself (separate from gas). These are typically included in the quoted swap price from aggregators.
Uniswap (Ethereum, Arbitrum, Base, Polygon, Optimism, BSC)
| Pool Tier | Fee | Typical Use |
|---|---|---|
| 0.01% | 1 bp | Stable-stable pairs (USDC/USDT) |
| 0.05% | 5 bps | Stable pairs, high-correlation pairs |
| 0.30% | 30 bps | Standard pairs (ETH/USDC, WBTC/ETH) |
| 1.00% | 100 bps | Exotic / low-liquidity pairs |
Curve Finance (Ethereum, Arbitrum, Polygon, others)
| Pool Type | Fee |
|---|---|
| Stablecoin pools | 0.01-0.04% |
| Crypto pools (v2) | 0.04-0.40% |
| Factory pools | Variable |
PancakeSwap (BSC, Ethereum, Arbitrum, Base)
| Pool Tier | Fee |
|---|---|
| V3 (stable) | 0.01% |
| V3 (standard) | 0.25% |
| V3 (exotic) | 1.00% |
| V2 | 0.25% |
Jupiter (Solana)
- Jupiter itself charges no protocol fee (aggregator layer)
- Underlying pool fees vary:
- Orca Whirlpools: 0.01%-2.00% (tier-based)
- Raydium: 0.25% (standard), 0.01% (concentrated)
- Meteora: 0.01%-1.00% (dynamic)
- Route optimization selects the best fee-adjusted path automatically
DEX Fee Notes
- Aggregator quotes already include DEX fees in the output amount. Do not double-count.
- When comparing CEX vs DEX prices, the DEX price is already net of protocol fees.
- The OKX DEX API aggregates across multiple DEX protocols and returns the best net quote.
- Some protocols have dynamic fees that adjust based on volatility.
14. Bridge Fee Tables
Typical fees for cross-chain asset transfers. Highly variable by bridge, route, and conditions.
Major Bridge Protocols
| Bridge | Typical Fee Range | Speed | Notes |
|---|---|---|---|
| OKX Bridge (via DEX API) | 0-0.1% | 1-15 min | Aggregates multiple bridges |
| Across Protocol | 0.04-0.12% | 1-5 min | Fast, uses relayers |
| Stargate | 0.06% | 5-15 min | LayerZero-based |
| Hop Protocol | 0.05-0.20% | 2-10 min | Uses AMMs |
| Synapse | 0.05-0.15% | 5-15 min | Multi-chain AMM |
| Native Bridges | Free (gas only) | 7 min - 7 days | Slowest, cheapest |
Native Bridge Wait Times
| Route | Deposit Time | Withdrawal Time |
|---|---|---|
| Ethereum -> Arbitrum | ~10 min | ~7 days (challenge period) |
| Ethereum -> Optimism | ~10 min | ~7 days (challenge period) |
| Ethereum -> Base | ~10 min | ~7 days (challenge period) |
| Ethereum -> Polygon | ~10 min | ~30 min (PoS bridge) |
| Ethereum -> BSC | N/A | N/A (use 3rd party) |
Bridge Fee Notes
- OKX withdrawal is often cheaper than bridging for moving assets from CEX to another chain. Withdraw directly to the target chain when supported.
- Bridge fees include: protocol fee + gas on source chain + gas on destination chain.
- For large amounts (>$100K), verify bridge liquidity on the destination chain to avoid high slippage.
- When possible, avoid bridging altogether by using native CEX withdrawal to the target chain.
CEX Withdrawal vs Bridge Comparison
When moving USDT from Ethereum to Arbitrum:
| Method | Cost | Time |
|---|---|---|
| OKX withdraw to Arbitrum directly | 0.1 USDT | 1-5 min |
| Bridge via Across | ~$2-5 (0.04-0.1%) + gas | 1-5 min |
| Bridge via native bridge | Gas only (~$15-30) | ~10 min + 7 days |
Recommendation: Always prefer direct CEX withdrawal to target chain when possible. It is almost always the cheapest and fastest route.
15. Cost Summary for Common Trade Patterns
Pattern A: CEX-DEX Arbitrage (Buy CEX, Sell DEX on Arbitrum)
Trade size: $50,000
CEX buy (VIP1 taker): $35.00 (7.0 bps)
CEX withdrawal (ETH to Arb): $0.34 (0.0001 ETH)
DEX sell gas (Arbitrum): $0.30
DEX slippage: $10.00 (est. 2 bps)
------
Total one-way cost: $45.64 (9.1 bps)
Pattern B: Funding Harvest (Spot long + Perp short)
Position size: $100,000
Spot buy (VIP1 taker): $70.00 (7.0 bps)
Perp short (VIP1 taker): $45.00 (4.5 bps)
------
Entry cost: $115.00 (11.5 bps)
Exit cost (same): $115.00 (11.5 bps)
Round-trip: $230.00 (23.0 bps)
Pattern C: Basis Trade (Spot long + Futures short)
Position size: $100,000
Spot buy (VIP1 taker): $70.00 (7.0 bps)
Futures short (VIP1 taker): $45.00 (4.5 bps)
------
Entry cost: $115.00 (11.5 bps)
Exit at expiry: futures settle automatically, close spot:
Spot sell: $70.00 (7.0 bps)
------
Total cost: $185.00 (18.5 bps)
16. Supported Chains Table
| Chain | chainIndex (OKX) | Native Token | Native Token Address (DEX) |
|---|---|---|---|
| Ethereum | 1 | ETH | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| BSC | 56 | BNB | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| Polygon | 137 | MATIC | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| Arbitrum One | 42161 | ETH | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| Optimism | 10 | ETH | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| Base | 8453 | ETH | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| Avalanche C-Chain | 43114 | AVAX | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
| Solana | 501 | SOL | 11111111111111111111111111111111 |
| X Layer | 196 | OKB | 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee |
Critical Warnings
Solana Native Token:
11111111111111111111111111111111= native SOL (System Program ID) -- use thisSo11111111111111111111111111111111111111112= Wrapped SOL (wSOL) -- do NOT use for native
EVM Native Token: All EVM chains use 0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee as the placeholder for native tokens.
Address Formatting:
- EVM: lowercase required
- Solana: base58, case-sensitive
BSC USDT Warning: BSC USDT has 18 decimals, unlike most other chains where it has 6 decimals. Always check the token's decimals value before encoding amounts.
17. Safety Checks
Profitability Safety Checks (#10-11 from Pre-Trade Checklist)
| # | Check | Tool | BLOCK Threshold | WARN Threshold | Error Code |
|---|---|---|---|---|---|
| 10 | Gas vs profit ratio | onchain-gateway gas + calculation |
Gas cost > 30% of gross spread | Gas cost > 10% of spread | NOT_PROFITABLE |
| 11 | Net profitability | profitability-calculator |
net_profit <= 0 |
profit_to_cost_ratio < 2 |
NOT_PROFITABLE |
Full Pre-Trade Safety Checklist (context)
Every skill runs through these checks in order before producing a recommendation:
| # | Check | BLOCK Threshold | WARN Threshold | Error Code |
|---|---|---|---|---|
| 1 | MCP connectivity | Server not reachable | -- | MCP_NOT_CONNECTED |
| 2 | Authentication | authenticated: false |
-- | AUTH_FAILED |
| 3 | Data freshness | > 60s stale (> 5s for arb) | > 30s stale | DATA_STALE |
| 4 | Token honeypot | is_honeypot === "1" |
-- | SECURITY_BLOCKED |
| 5 | Token tax rate | buy > 5% OR sell > 10% | buy > 1% | SECURITY_BLOCKED |
| 6 | Holder concentration | Top 10 non-contract > 80% | > 50% | SECURITY_BLOCKED |
| 7 | Contract verified | is_open_source === "0" |
-- | SECURITY_BLOCKED |
| 8 | Liquidity depth | liquidityUsd < $100,000 |
< $500,000 |
INSUFFICIENT_LIQUIDITY |
| 9 | Price impact | priceImpactPercent > 2% |
> 0.5% |
INSUFFICIENT_LIQUIDITY |
| 10 | Gas vs profit ratio | Gas > 30% of spread | Gas > 10% of spread | NOT_PROFITABLE |
| 11 | Net profitability | net_profit <= 0 |
profit_to_cost_ratio < 2 |
NOT_PROFITABLE |
BLOCK Conditions
| Condition | Action | Error Code |
|---|---|---|
net_profit <= 0 |
Output [NOT PROFITABLE] with full breakdown. Do not recommend. |
NOT_PROFITABLE |
total_costs_usd cannot be computed (missing data) |
Output error with details. | FEE_LOOKUP_FAILED |
| Required field missing in TradeLeg | Output validation error. | MISSING_FIELD |
| Invalid field value | Output validation error. | INVALID_LEG |
WARN Conditions
| Condition | Warning Message |
|---|---|
profit_to_cost_ratio < 2.0 |
"利潤率偏低，執行風險大 / Thin margin, execution risk is high" |
gas_cost > 30% of gross_spread |
"Gas 費佔毛利 {pct}%，考慮使用 L2 / Gas is {pct}% of gross, consider L2" |
estimated_slippage > 100 bps (1%) |
"滑點估計偏高 ({bps} bps)，減小規模或分批執行 / High slippage, reduce size or split" |
withdrawal_fee > 10% of net_profit |
"提幣費佔淨利 {pct}%，考慮直接提到目標鏈 / Withdrawal fee is {pct}% of profit" |
confidence == "low" |
"數據來源不完整，結果僅供參考 / Incomplete data sources, treat as estimate only" |
hold_days > 7 and funding_rate used |
"長期持有假設固定資金費率，實際會波動 / Long hold assumes constant funding, actual varies" |
18. Error Codes
| Code | Condition | User Message (ZH) | User Message (EN) |
|---|---|---|---|
INVALID_LEG |
Leg field has invalid value | 交易腿參數無效：{field} = {value} | Invalid trade leg parameter: {field} = {value} |
MISSING_FIELD |
Required field missing | 缺少必要參數：{field}（{leg_description}） | Missing required field: {field} ({leg_description}) |
FEE_LOOKUP_FAILED |
Cannot determine fee rate | 費率查詢失敗：{detail}。請確認 VIP 等級和交易類型。 | Fee lookup failed: {detail}. Verify VIP tier and instrument type. |
GAS_FETCH_FAILED |
Cannot get live gas price | Gas 價格獲取失敗，使用基準值估算（信心度降為 medium） | Gas price fetch failed. Using benchmark estimate (confidence: medium). |
ORDERBOOK_EMPTY |
Orderbook returned no levels | 訂單簿為空或流動性不足：{instId} | Orderbook empty or insufficient liquidity: {instId} |
NOT_PROFITABLE |
Net profit <= 0 | 扣除所有成本後淨利潤為負（{net_pnl}），不建議執行 | Net profit is negative after all costs ({net_pnl}). Not recommended. |
MARGIN_TOO_THIN |
Profit-to-cost < 2.0 | 利潤空間偏薄（利潤/成本比 = {ratio}x），風險較高 | Thin margin (profit/cost = {ratio}x). Higher execution risk. |
MCP_NOT_CONNECTED |
MCP server unreachable | MCP 伺服器無法連線，請檢查 okx-trade-mcp 是否啟動 | MCP server unreachable. Check if okx-trade-mcp is running. |
AUTH_FAILED |
API key invalid or expired | API 認證失敗，請檢查 OKX API 金鑰設定 | API authentication failed. Check OKX API key configuration. |
DATA_STALE |
Price/market data too old | 市場數據已過期（超過 {threshold} 秒），正在重新獲取... | Market data stale (> {threshold}s). Refetching... |
RATE_LIMITED |
API rate limit hit | API 請求頻率超限，{wait}秒後重試 | API rate limit reached. Retrying in {wait}s. |
19. Cross-Skill Integration
Input Contract: Who Calls This Skill
| Calling Skill | When | Data Provided |
|---|---|---|
cex-dex-arbitrage |
After finding spread opportunity | TradeLeg[] with CEX + DEX legs, prices pre-filled |
funding-rate-arbitrage |
After finding funding opportunity | TradeLeg[] with spot + swap legs, funding_rate + intervals |
basis-trading |
After finding basis opportunity | TradeLeg[] with spot + futures legs, hold_days |
yield-optimizer |
When computing switching costs | TradeLeg[] representing the rebalance (exit old + enter new) |
smart-money-tracker |
Before recommending copy trade | TradeLeg[] for the proposed copy |
| User (direct) | Interactive cost analysis | TradeLeg[] manually constructed |
Output Contract: What This Skill Returns
All callers receive a ProfitabilityResult (see Section 7.1).
| Consuming Skill | Fields Used | Decision Logic |
|---|---|---|
cex-dex-arbitrage |
is_profitable, net_profit_usd, profit_to_cost_ratio |
Skip if is_profitable == false or profit_to_cost_ratio < 1.5 |
funding-rate-arbitrage |
total_costs_usd (as entry cost), min_spread_for_breakeven_bps |
Compute break-even holding days: total_costs / daily_income |
basis-trading |
total_costs_usd, cost_breakdown |
Subtract from basis yield to get net yield |
yield-optimizer |
total_costs_usd (switching cost) |
Compute break-even period for yield switch |
smart-money-tracker |
is_profitable, warnings |
Gate copy recommendation |
Data Flow Diagram
price-feed-aggregator.snapshot
│
│ PriceSnapshot[] (CEX + DEX prices)
▼
[strategy skill].scan / evaluate
│
│ Constructs TradeLeg[] with prices
▼
profitability-calculator.estimate
│
│ ProfitabilityResult
▼
[strategy skill] applies go/no-go decision
│
▼
Output to user (with cost breakdown)
20. Output Templates
20.1 Estimate / Breakdown Output
══════════════════════════════════════════
[DEMO] [RECOMMENDATION ONLY -- 不會自動執行]
══════════════════════════════════════════
## 盈利性分析 / Profitability Analysis
策略: CEX 買入 -> DEX 賣出 (ETH on Ethereum)
規模: $1,000.00
VIP 等級: VIP0
── 成本分解 ──────────────────────────────
| 項目 | 金額 | 佔毛利 |
|------|------|--------|
| 毛利差 (Gross Spread) | +$17.77 | 100.0% |
| CEX Taker 費 (0.08%) | -$0.80 | 4.5% |
| DEX Gas 費 (Ethereum) | -$5.20 | 29.3% |
| DEX 滑點 (est. 12 bps) | -$1.20 | 6.8% |
| OKX 提幣費 (ETH ERC-20) | -$1.55 | 8.7% |
| **總成本** | **-$8.75** | **49.2%** |
| **淨利潤** | **+$9.02** | **50.8%** |
利潤/成本比: 1.03x
最小盈利價差: 88 bps
信心度: HIGH
結果: [PROFITABLE]
注意:
- Gas 費佔毛利 29.3%，接近 30% 閾值
考慮使用 Layer 2 (Arbitrum/Base) 降低 Gas 成本
- 利潤/成本比 1.03x 低於 2.0x 建議閾值
執行風險較大，需精確時機
══════════════════════════════════════════
Next Steps
══════════════════════════════════════════
1. 考慮切換至 Arbitrum 降低 Gas 成本:
profitability-calculator estimate --legs '[...chain:arbitrum...]'
2. 查看不同規模的敏感度分析:
profitability-calculator sensitivity --legs '[...]' --variable size_usd --range '[500,5000]'
3. 查看最低價差要求:
profitability-calculator min-spread --venue-a okx-cex --venue-b okx-dex --chain ethereum --asset ETH --size-usd 1000
── Disclaimer ────────────────────────────
This is analysis only. No trades are executed automatically.
All recommendations require manual review and execution.
以上僅為分析建議，不會自動執行任何交易。
所有建議均需人工審核後手動操作。
══════════════════════════════════════════
20.2 Min-Spread Output
══════════════════════════════════════════
[DEMO] [RECOMMENDATION ONLY -- 不會自動執行]
══════════════════════════════════════════
## 最小價差分析 / Minimum Spread Analysis
路線: OKX (CEX) <-> OKX DEX (Ethereum)
資產: ETH | 規模: $10,000.00 | VIP: VIP0
── 成本組成 (bps) ────────────────────────
| 成本項目 | bps |
|----------|-----|
| CEX Taker 費 | 8.0 |
| DEX Gas 費 | 1.6 |
| 滑點估計 | 3.0 |
| 提幣費 | 1.2 |
| **合計 (最小價差)** | **13.8** |
結論: 至少需要 14 bps 的價差才能盈虧平衡 ($10,000 規模)。
══════════════════════════════════════════
20.3 Sensitivity Output
══════════════════════════════════════════
[DEMO] [RECOMMENDATION ONLY -- 不會自動執行]
══════════════════════════════════════════
## 敏感度分析 / Sensitivity Analysis
變量: size_usd (交易規模)
基礎策略: CEX 買入 -> DEX 賣出 (ETH, Arbitrum)
| 規模 | 總成本 | 淨利潤 | 利潤/成本 | 盈利? |
|------|--------|--------|-----------|-------|
| $1,000 | $1.55 | +$0.59 | 0.38x | [WARN] |
| $3,000 | $2.95 | +$3.47 | 1.18x | [OK] |
| $5,000 | $4.35 | +$6.35 | 1.46x | [OK] |
| $10,000 | $7.50 | +$14.20 | 1.89x | [OK] |
| $20,000 | $13.80 | +$29.60 | 2.14x | [GOOD] |
| $50,000 | $33.00 | +$75.70 | 2.29x | [GOOD] |
盈虧平衡點: ~$620
最佳利潤/成本比: $50,000 (2.29x)
結論: 規模越大，固定成本（Gas、提幣費）攤薄越多。
建議最小規模: $3,000+（利潤/成本比 > 1.0x）
══════════════════════════════════════════
21. Conversation Examples
Example 1: Direct Profitability Check (CEX-DEX Arbitrage)
User: ETH OKX 現貨 3412.50，Uniswap 上 3419.80，$5000 規模，賺唔賺？
Assistant Flow:
1. Parse: Buy OKX CEX at 3412.50, Sell Uniswap (Ethereum) at 3419.80, $5,000
2. Construct TradeLeg[]:
Leg 1: {venue: "okx-cex", asset: "ETH", side: "buy", size_usd: 5000, order_type: "market"}
Leg 2: {venue: "uniswap", chain: "ethereum", asset: "ETH", side: "sell", size_usd: 5000,
withdrawal_required: true}
3. Fetch live data:
- market_get_orderbook(instId="ETH-USDT", sz="400")
→ Walk asks to estimate slippage for $5,000 buy
→ Slippage: ~2 bps
- onchainos onchain-gateway gas --chain ethereum
→ gasPriceGwei: "28"
- Gas limit benchmark: 200,000 (Uniswap swap)
- Native token price: $3,412.50 (from ticker)
- Withdrawal fee: 0.00035 ETH = $1.19
4. Compute:
gross_spread = (3419.80 - 3412.50) / 3412.50 * 5000 = $10.69
cex_trading_fee = 5000 * 0.0008 = $4.00
gas_cost = 28 * 200000 * 3412.50 / 1e9 = $19.11
slippage_cost = 5000 * 2 / 10000 = $1.00
withdrawal_fee = 0.00035 * 3412.50 = $1.19
total_costs = 4.00 + 19.11 + 1.00 + 1.19 = $25.30
net_profit = 10.69 - 25.30 = -$14.61
5. Result: NOT PROFITABLE
Output:
══════════════════════════════════════════
[DEMO] [RECOMMENDATION ONLY -- 不會自動執行]
══════════════════════════════════════════
## 盈利性分析
策略: OKX 買入 -> Uniswap 賣出 (ETH on Ethereum)
規模: $5,000.00
── 成本分解 ──────────────────────────────
| 項目 | 金額 | 佔毛利 |
|------|------|--------|
| 毛利差 | +$10.69 | 100.0% |
| CEX Taker 費 (0.08%) | -$4.00 | 37.4% |
| DEX Gas 費 (28 gwei) | -$19.11 | 178.8% |
| CEX 滑點 (2 bps) | -$1.00 | 9.4% |
| 提幣費 (ETH ERC-20) | -$1.19 | 11.1% |
| **總成本** | **-$25.30** | **236.7%** |
| **淨利潤** | **-$14.61** | -- |
結果: [NOT PROFITABLE]
主要原因: Ethereum Gas 費 ($19.11) 遠超毛利 ($10.69)。
建議:
1. 改用 Arbitrum: Gas 約 $0.30，預估淨利 +$4.20
2. 增大規模至 $20,000+: 攤薄固定成本
3. 等待 Gas 降至 <10 gwei 再行動
══════════════════════════════════════════
Example 2: Funding Rate Carry Trade Cost Analysis
User: I want to do ETH funding harvest, $50,000 position, VIP1. What's the cost to enter and exit?
Assistant Flow:
1. Construct TradeLeg[] for entry:
Leg 1: {venue: "okx-cex", asset: "ETH", side: "buy", size_usd: 50000,
instrument_type: "spot", vip_tier: "VIP1"}
Leg 2: {venue: "okx-cex", asset: "ETH", side: "sell", size_usd: 50000,
instrument_type: "swap", vip_tier: "VIP1"}
2. Compute entry cost:
spot_buy_fee = 50000 * 0.0007 = $35.00 (VIP1 spot taker)
swap_sell_fee = 50000 * 0.00045 = $22.50 (VIP1 swap taker)
slippage (spot): fetch orderbook → ~1 bps → $5.00
slippage (swap): fetch orderbook → ~1 bps → $5.00
entry_cost = 35.00 + 22.50 + 5.00 + 5.00 = $67.50
3. Exit cost is symmetric:
exit_cost = $67.50
4. Round-trip total: $135.00
5. Fetch current funding rate:
market_get_funding_rate(instId="ETH-USDT-SWAP")
→ fundingRate: "0.000150" (0.015% per 8h)
daily_income = 50000 * 0.000150 * 3 = $22.50
break_even_days = 135.00 / 22.50 = 6.0 days
Output:
══════════════════════════════════════════
[DEMO] [RECOMMENDATION ONLY -- 不會自動執行]
══════════════════════════════════════════
## 成本分析: ETH 資金費率套利
規模: $50,000.00 | VIP: VIP1
── 建倉成本 ──────────────────────────────
| 項目 | 金額 | bps |
|------|------|-----|
| 現貨買入手續費 (VIP1 taker) | $35.00 | 7.0 |
| 永續合約做空手續費 (VIP1 taker) | $22.50 | 4.5 |
| 現貨滑點 (est. 1 bps) | $5.00 | 1.0 |
| 永續滑點 (est. 1 bps) | $5.00 | 1.0 |
| **建倉總成本** | **$67.50** | **13.5** |
── 平倉成本 (對稱) ───────────────────────
| 項目 | 金額 | bps |
|------|------|-----|
| **平倉總成本** | **$67.50** | **13.5** |
── 總結 ──────────────────────────────────
| 指標 | 數值 |
|------|------|
| 總進出成本 | $135.00 (27.0 bps) |
| 當前資金費率 | +0.015% / 8h (16.43% ann.) |
| 每日預估收入 | +$22.50 |
| 盈虧平衡天數 | 6.0 天 |
結果: [PROFITABLE] (持有 > 6 天後)
注意: 資金費率每 8 小時變動一次，以上預估假設費率不變。
建議持續監控費率走勢。
══════════════════════════════════════════
Example 3: Min-Spread with Chain Comparison
User: ETH 在 OKX 和 DEX 之間最少要幾 bps 價差先有得賺？$10,000，比較 Ethereum 同 Arbitrum。
Assistant Flow:
1. Run min-spread for Ethereum:
profitability-calculator min-spread --venue-a okx-cex --venue-b okx-dex
--chain ethereum --asset ETH --size-usd 10000
CEX taker (VIP0): 8.0 bps → $8.00
Gas (Ethereum, 30 gwei, 200k limit): $20.70 → 20.7 bps
Slippage: ~2 bps → $2.00
Withdrawal (ETH ERC-20): $1.19 → 1.2 bps
Total: 31.9 bps
2. Run min-spread for Arbitrum:
profitability-calculator min-spread --venue-a okx-cex --venue-b okx-dex
--chain arbitrum --asset ETH --size-usd 10000
CEX taker (VIP0): 8.0 bps → $8.00
Gas (Arbitrum, 0.3 gwei, 1.5M limit): $1.55 → 1.6 bps
Slippage: ~2 bps → $2.00
Withdrawal (ETH Arbitrum): $0.35 → 0.4 bps
Total: 12.0 bps
Output:
══════════════════════════════════════════
[DEMO] [RECOMMENDATION ONLY -- 不會自動執行]
══════════════════════════════════════════
## 最小價差比較: Ethereum vs Arbitrum
資產: ETH | 規模: $10,000.00 | VIP: VIP0
── Ethereum ──────────────────────────────
| 成本項目 | bps | USD |
|----------|-----|-----|
| CEX Taker 費 | 8.0 | $8.00 |
| Gas 費 (30 gwei) | 20.7 | $20.70 |
| 滑點估計 | 2.0 | $2.00 |
| 提幣費 (ERC-20) | 1.2 | $1.19 |
| **合計** | **31.9** | **$31.89** |
── Arbitrum ──────────────────────────────
| 成本項目 | bps | USD |
|----------|-----|-----|
| CEX Taker 費 | 8.0 | $8.00 |
| Gas 費 (0.3 gwei) | 1.6 | $1.55 |
| 滑點估計 | 2.0 | $2.00 |
| 提幣費 (Arbitrum) | 0.4 | $0.35 |
| **合計** | **12.0** | **$11.90** |
── 比較 ──────────────────────────────────
| 鏈 | 最小價差 | 節省 |
|----|---------|------|
| Ethereum | 32 bps | -- |
| Arbitrum | 12 bps | -20 bps (62%) |
結論: Arbitrum 所需最小價差僅為 Ethereum 的 38%。
Gas 費是主要差異來源（20.7 bps vs 1.6 bps）。
強烈建議使用 Arbitrum 執行 ETH CEX-DEX 套利。
══════════════════════════════════════════
22. Implementation Notes
Orderbook Depth Parsing
When auto-calculating slippage from market_get_orderbook:
1. Call: market_get_orderbook(instId="ETH-USDT", sz="400")
2. Parse bids (for sell) or asks (for buy) - remember all values are STRINGS
3. Convert size_usd to asset quantity: qty = size_usd / best_price
4. Walk levels:
remaining = qty
weighted_sum = 0
for level in levels:
level_price = parseFloat(level[0])
level_qty = parseFloat(level[1])
fill = min(remaining, level_qty)
weighted_sum += level_price * fill
remaining -= fill
if remaining <= 0: break
5. vwap = weighted_sum / qty
6. impact_bps = abs(vwap - best_price) / best_price * 10000
Fee Rate Lookup Priority
1. If user provides explicit fee_rate → use it
2. If vip_tier + instrument_type provided → lookup from fee tables above
3. Default: VIP0 taker rate for the instrument type
- Spot: 0.080%
- Swap/Futures: 0.050%
23. Changelog
| Date | Version | Changes |
|---|---|---|
| 2026-03-09 | 1.0.0 | Initial release. 4 commands: estimate, breakdown, min-spread, sensitivity. |
| 2026-03-09 | 1.1.0 | Inlined all reference file content for OpenClaw/Lark compatibility. |
Note: All content from reference files has been inlined above for OpenClaw/Lark compatibility. The files below exist for human maintenance only.
| File | Relevance to This Skill |
|---|---|
references/formulas.md |
Calculation formulas (Sections 1-2, 7) |
references/fee-schedule.md |
OKX fee tiers, withdrawal fees, gas benchmarks, DEX fees, bridge fees |
references/chain-reference.md |
Supported chains, native addresses, chainIndex mapping |
references/safety-checks.md |
Pre-trade safety checklist, profitability checks #10-11 |
Weekly Installs
1
Repository
foxisyw/skills-…x-stratsFirst Seen
8 days ago
Security Audits
Installed on
amp1
cline1
openclaw1
opencode1
cursor1
kimi-cli1