Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Tencent SkillHub · AI
OKX Exchange
OKX quantitative trading agent. Supports spot, USDT-M perpetual swap, and coin-M futures. Strategies: grid trading, trend following (MA/RSI/MACD), spot-futur...
skill openclawclawhub Free
OKX quantitative trading agent. Supports spot, USDT-M perpetual swap, and coin-M futures. Strategies: grid trading, trend following (MA/RSI/MACD), spot-futur...
⬇ 0 downloads ★ 0 stars Unverified but indexed
Install for OpenClaw
Quick setup
- Download the package from Yavira.
- Extract the archive and review SKILL.md first.
- Import or place the package into your OpenClaw setup.
Package facts
- Download mode
- Yavira redirect
- Package format
- ZIP package
- Source platform
- Tencent SkillHub
- What's included
- README.md, README_ZH.md, SKILL.md, docs/decision-engine-guide.md, docs/learning-system-data-management.md, docs/trading-rules.md
Validation
- Use the Yavira download entry.
- Review SKILL.md after the package is downloaded.
- Confirm the extracted package contains the expected setup assets.
Install with your agent
Agent handoff
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
New install
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete.
Upgrade existing
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run.
Trust & source
Release facts
- Source
- Tencent SkillHub
- Verification
- Indexed source record
- Version
- 1.0.1
Provenance
- Publisher
- lk2023060901
- Source page
- View original listing
- Canonical URL
- Open canonical page
Documentation
OKX Exchange Skill
Quantitative trading agent for OKX exchange via the official V5 API. API Docs: https://www.okx.com/docs-v5/en/
Setup (First Time)
# 1. Add credentials to .env cat >> ~/.openclaw/workspace/.env << 'EOF' OKX_API_KEY=your_key OKX_SECRET_KEY=your_secret OKX_PASSPHRASE=your_passphrase OKX_SIMULATED=1 EOF # 2. Run setup cd ~/.openclaw/workspace/skills/okx-exchange/scripts python setup.py Important: OKX_SIMULATED=1 = paper trading (safe). Set to 0 for live trading.
Session Init
On every session, load credentials first: source ~/.openclaw/workspace/.env cd ~/.openclaw/workspace/skills/okx-exchange/scripts
Unified CLI (Recommended)
Use okx.py as the single entry point for all commands: python3 okx.py help # Show all commands python3 okx.py account # Portfolio overview python3 okx.py buy BTC-USDT market 0.01 python3 okx.py trend analyze BTC-USDT-SWAP All individual scripts remain usable directly, but okx.py is preferred for agent use.
Instrument ID Format
TypeFormatExampleSpot{BASE}-{QUOTE}BTC-USDTPerpetual Swap{BASE}-{QUOTE}-SWAPBTC-USDT-SWAPFutures{BASE}-{QUOTE}-{EXPIRY}BTC-USDT-250328 tdMode: cash (spot), cross (cross-margin), isolated (isolated-margin)
Account & Portfolio
python3 okx.py account # Full portfolio summary python3 okx.py account balance # Balances only python3 okx.py account balance USDT # Specific currency python3 okx.py account positions # All positions python3 okx.py account orders # Pending orders python3 okx.py account history # Filled order history (SPOT) python3 okx.py account history SWAP # Perpetual swap order history python3 okx.py account history SPOT BTC-USDT # History for specific symbol
Trading Mode (Live / Demo)
python3 okx.py mode # Show current mode python3 okx.py mode demo # Switch to paper trading (safe) python3 okx.py mode live # Switch to live trading (requires confirmation + LIVE credentials) Live credentials use separate env vars: OKX_API_KEY_LIVE / OKX_SECRET_KEY_LIVE / OKX_PASSPHRASE_LIVE
Order Execution
# Spot market buy python3 okx.py buy BTC-USDT market 0.01 # Spot limit sell python3 okx.py sell BTC-USDT limit 0.01 --px 50000 # Perpetual swap — long with TP/SL (attached to the order) python3 okx.py buy BTC-USDT-SWAP market 1 --td cross --pos long --tp 55000 --sl 42000 # Perpetual swap — short python3 okx.py sell BTC-USDT-SWAP market 1 --td cross --pos short # Close position (reduce-only) python3 okx.py sell BTC-USDT-SWAP market 1 --td cross --pos long --reduce # Cancel order python3 okx.py cancel BTC-USDT <ord_id> # Cancel all orders for symbol python3 okx.py cancel-all BTC-USDT # Set leverage python3 okx.py leverage BTC-USDT-SWAP 10 --td cross # Transfer funds between accounts (6=Funding, 18=Trading) python3 okx.py transfer USDT 500 funding trading # Funding → Trading python3 okx.py transfer USDT 500 trading funding # Trading → Funding Skip confirmation prompt (for automation): python3 okx.py buy BTC-USDT market 0.01 --no-confirm
Algo Orders (Standalone TP/SL — for existing positions)
Use when a position is already open and you want to add a stop-loss or take-profit separately. Unlike --tp/--sl flags (which attach to the entry order), algo orders are independent. # OCO: TP + SL together — whichever triggers first cancels the other python3 okx.py algo oco BTC-USDT-SWAP 1 --tp 55000 --sl 45000 --td cross --pos long --reduce # Conditional: single stop-loss only python3 okx.py algo stop BTC-USDT-SWAP 1 --sl 45000 --td cross --reduce # List all pending algo orders python3 okx.py algo list python3 okx.py algo list BTC-USDT-SWAP # filter by instrument # Cancel a specific algo order python3 okx.py algo cancel BTC-USDT-SWAP <algo_id> When to use which: --tp/--sl at order time → atomic, guaranteed to attach, best for new entries algo oco → add/replace TP+SL after entry, or adjust levels mid-trade
Trend Following (MA + RSI + MACD)
# Analyze only (no trade) python3 okx.py trend analyze BTC-USDT-SWAP --bar 1H # Run with auto-execution python3 okx.py trend run BTC-USDT-SWAP 1 --bar 4H --tp 0.05 --sl 0.03 --td cross --pos long # Dry run (analysis only, no trade) python3 okx.py trend run BTC-USDT-SWAP 1 --dry Signal logic: BUY: MA_fast > MA_slow + RSI < 70 + MACD histogram > 0 SELL: MA_fast < MA_slow + RSI > 30 + MACD histogram < 0
Grid Trading
# Setup grid: BTC-USDT, range 40000-50000, 10 grids, 1000 USDT total python3 okx.py grid setup BTC-USDT 40000 50000 10 1000 # Check and rebalance filled orders (run periodically) python3 okx.py grid check BTC-USDT # Stop and cancel all grid orders python3 okx.py grid stop BTC-USDT
Spot-Futures Arbitrage
# Scan all pairs for opportunities python3 okx.py arb scan # Check specific pair basis python3 okx.py arb basis BTC-USDT BTC-USDT-SWAP # Open arbitrage position (buy spot + short swap) python3 okx.py arb open BTC-USDT BTC-USDT-SWAP 1000 --min-basis 0.1 # Close arbitrage position when basis converges python3 okx.py arb close BTC-USDT BTC-USDT-SWAP 0.01 1 --max-basis 0.02
Snapshot Report
Fetches live data from the API on every run, persists the snapshot, and outputs a full report with a real historical tracking table. python3 okx.py snapshot Output includes: Total equity, initial capital, % change (relative to first snapshot) Available USDT, major holdings Contract positions (entry price, mark price, unrealized PnL, liquidation distance) Historical tracking table (from memory/okx-monitor-snapshots.json, up to 48 entries) All figures are sourced directly from the OKX API — no estimates or fabricated values.
Automated Monitoring (Cron)
# Default: sl-tp every 5 minutes, scan every 30 minutes bash scripts/cron_setup.sh setup # Custom intervals (dynamic parameters) bash scripts/cron_setup.sh setup 1m # sl-tp every 1 minute bash scripts/cron_setup.sh setup 10m 1h # sl-tp every 10 minutes, scan every 1 hour # Stop all jobs bash scripts/cron_setup.sh teardown # Show status bash scripts/cron_setup.sh status Cron jobs: okx-sl-tp — periodically runs snapshot + monitor sl-tp, reports live data okx-scan — periodically runs monitor scan, reports strategy signals
Adjust cron interval (via user instruction)
If the user asks to change the monitoring frequency (e.g. "change to every 10 minutes"): # Find the cron job ID first openclaw cron list # Then edit the interval (replace <id> with the okx-sl-tp job ID) openclaw cron edit <id> --every 10m # e.g. 1m, 5m, 15m, 30m, 1h
Manual monitor run
python3 okx.py snapshot # Full account snapshot report (with history — recommended) python3 okx.py monitor # SL/TP check + strategy scan + liquidation risk python3 okx.py monitor sl-tp # Only SL/TP check python3 okx.py monitor scan # Only strategy scan python3 okx.py monitor liq-risk # Liquidation risk check (default: alert if within 10%) python3 okx.py monitor liq-risk 5 # Alert if mark price within 5% of liquidation price
CRITICAL: Cron reporting rules
Every cron report MUST call the scripts first. NEVER fabricate, estimate, or reuse previous numbers. Recommended single command for cron: cd ~/.openclaw/workspace/skills/okx-exchange/scripts python3 okx.py snapshot python3 okx.py monitor sl-tp snapshot automatically saves the snapshot and generates a report with historical tracking. Forward the output verbatim — do not add, remove, or rephrase any content. Report only what the scripts actually output. If a script returns an error, report the error — do not substitute with previously seen values. Why this matters: Each cron run is an isolated session with no memory of previous runs. Any numbers not fetched from the API in this session are either fabricated or stale.
Performance Report
python3 okx.py report daily # Today's P&L summary python3 okx.py report weekly # Last 7 days python3 okx.py report all # All-time Output includes: trade count, win rate, total P&L, best/worst trade, and top-5 coin breakdown.
Configuration
View or update preferences without editing JSON directly: python3 okx.py prefs show python3 okx.py prefs set auto_trade true python3 okx.py prefs set stop_loss_pct 3.0 python3 okx.py prefs set watchlist BTC-USDT-SWAP,ETH-USDT-SWAP,SOL-USDT-SWAP Full preferences schema (~/.openclaw/workspace/memory/okx-trading-preferences.json): { "max_order_usd": 100, "max_leverage": 10, "price_impact_warn": 0.005, "price_impact_abort": 0.01, "require_confirm": true, "stop_loss_pct": 5.0, "take_profit_pct": 10.0, "auto_trade": false, "max_position_usd": 100, "max_daily_trades": 10, "default_sz": "0.01", "strategies": ["trend"], "watchlist": ["BTC-USDT-SWAP", "ETH-USDT-SWAP"] } ParameterDefaultDescriptionmax_order_usd100Max USD per single ordermax_leverage10Max allowed leverageprice_impact_warn0.005Warn if market order impact > 0.5%price_impact_abort0.01Abort if market order impact > 1%require_confirmtruePrompt before placing ordersstop_loss_pct5.0Auto stop-loss at -X% unrealized PnLtake_profit_pct10.0Auto take-profit at +X% unrealized PnLauto_tradefalseExecute trades automatically (no confirmation)max_position_usd100Max USD per positionmax_daily_trades10Max auto-trades per daydefault_sz"0.01"Default size for auto-tradingstrategies["trend"]Active strategies: trend / arbitragewatchlist[...]Instruments for monitor scan
Environment Variables
VariableDefaultDescriptionOKX_API_KEY—Demo API keyOKX_SECRET_KEY—Demo API secretOKX_PASSPHRASE—Demo API passphraseOKX_API_KEY_LIVE—Live API key (used when mode=live)OKX_SECRET_KEY_LIVE—Live API secretOKX_PASSPHRASE_LIVE—Live API passphraseOKX_SIMULATED0Set 1 for paper trading (demo account)OKX_WS0Set 1 to enable WebSocket feed (real-time cache for tickers/candles/account/positions)OKX_API_URLhttps://www.okx.comOverride API base URLOKX_LOG_LEVELINFODEBUG / INFO / WARNING / ERROROKX_LOG_FORMATtexttext or json (structured logging)OKX_CRON_MODE0Set 1 to suppress INFO output (for cron)
WebSocket Mode (OKX_WS=1)
When enabled, a background thread maintains real-time caches: Public feed: tickers, candles — OKXClient.ticker() / candles() read from cache first Private feed: account balances, positions, orders — balance() / positions() read from cache first REST API is used as fallback when cache is empty or WS is disabled
Trading Rules
Before making any buy/sell decision, read and apply the principles in: docs/trading-rules.md These rules define how to observe the market, weigh signals, size positions, and learn from outcomes. They do not prescribe a fixed strategy — the agent is expected to adapt its approach based on what the market is actually showing.
Safety Rules
Always confirm before placing orders unless auto_trade=true or --no-confirm Price impact check: abort if >1%, warn if >0.5% Start with OKX_SIMULATED=1 — validate everything before going live Never print or log credentials — load via source .env only Set stop-loss on every position — meme coins and leveraged positions can move fast Daily trade limit — auto-trading will stop when max_daily_trades is reached
Memory Files
FilePurposememory/okx-trading-preferences.jsonStrategy config, risk parameters, and trading modememory/okx-trading-state.jsonRuntime state (daily trade count, last scan)memory/okx-trading-journal.jsonTrade history from monitor (SL/TP closures, auto-entries)memory/okx-trade-journal.jsonTrade history from learning system (signal analysis)memory/okx-learning-model.jsonLearning model (win rates by coin/regime, optimal params)memory/okx-grid-{inst_id}.jsonGrid state per instrument
Kill Switch
# Stop all automated trading bash scripts/cron_setup.sh teardown # Or remove OKX_API_KEY from .env to prevent any API calls
Category context
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Source: Tencent SkillHub
Largest current source with strong distribution and engagement signals.
Package contents
Included in package
6 Docs
- SKILL.md
- docs/decision-engine-guide.md
- docs/learning-system-data-management.md
- docs/trading-rules.md
- README_ZH.md
- README.md