wy51ai/web3-starter-kit/web3-trade-simulator
- community
- ext:[{"type":"plugin","description":"Custom plugin integration"}]
Checking that the SKILL.md file is still in the repository.
Use this skill when the user wants to practice trading without real money, or evaluate their trading ability. Triggers: '模拟买入', '纸上交易', '假设我买了', '帮我练习交易', '我的模拟仓位', '模拟卖出', '我的交易能力怎么样', '帮我复盘', 'paper trade', 'simulate a trade', 'what if I had bought', 'test my trading skills', 'track my virtual portfolio', 'score my trades', 'how am I doing'. This skill uses real on-chain data from onchainos CLI to simulate realistic trade outcomes, score decision quality, and teach beginners through hands-on practice. Always respond in the same language the user uses (Chinese or English).
Install
npx vskill@latest install wy51ai/web3-starter-kit/web3-trade-simulatorSemi-extensibleMentions customization, but needs skill-specific knowledge. Read the extensible skills standard
SKILL.md
Web3 Trade Simulator — 模拟交易训练营
Lets Web3 beginners practice trading with zero real money at risk, using live on-chain data from the onchainos CLI. Every simulated trade is scored across five dimensions so users learn why good trades work.
Prerequisites
Requires the onchainos CLI (same as web3-safe-guide):
Core Principles
- Real data, zero risk: All prices and market data come from live onchainos calls — the simulation is realistic, but no wallet is touched.
- Score every decision: After each trade, give an objective score so users learn, not just play.
- State lives in conversation: The virtual portfolio is maintained as a running ledger in the conversation. Be explicit when the user's session ends ("your virtual portfolio will reset if we start a new chat").
- Education woven in: Explain why a score is what it is. Never just say "bad trade" — say which factor dragged the score.
- Safety gate still applies: Honeypot tokens are blocked even in simulation. No point teaching bad habits.
Virtual Portfolio State
The portfolio is persisted locally so it survives across conversations.
Storage
- File:
~/.web3-trade-simulator/portfolio.json - On skill start: Read the file if it exists and resume from saved state. If the file is missing, initialize a fresh $10,000 portfolio and create the file.
- After every trade action (buy, sell, reset): Write the full updated state back to the file immediately.
JSON Schema
Ledger Display
Display this ledger at the start of every response when the user has at least one open or closed position.
Trade Decision Score (0–100)
Score each entry decision across five dimensions. Show breakdown after every simulated buy.
| Dimension | Weight | How to Compute | |
|---|---|---|---|
| Safety | 30 pts | Use full Safety Score from web3-safe-guide workflow. Map 0–100 → 0–30 pts linearly. | |
| Smart Money Alignment | 20 pts | Run signal check (below). Signal present + still holding: +20; signal faded/none: +0; signal already fully exited: −10. | |
| Trend Timing | 20 pts | Analyze last-12-candle kline. Buying into uptrend: +20; sideways: +10; buying into downtrend: +0. | |
| Liquidity Discipline | 15 pts | liquidity from price-info. >$1M: +15; $100K–$1M: +10; $10K–$100K: +5; <$10K: −15. | |
| Position Sizing | 15 pts | Evaluate the virtual amount vs current balance. ≤10% of balance: +15; 11–25%: +10; 26–50%: +5; >50%: −15. |
Score Levels:
Workflow 1 — Simulate a Buy
Triggers: "模拟买", "假设我买了X", "paper buy X", "simulate buying X"
Steps
Block Conditions
isHoneyPot = true→ "☠️ 即使是模拟交易,骗局代币也不在我们的训练内容里。我来给你展示一个更值得研究的代币。"- Safety Score < 30 → Warn strongly but allow if user confirms ("这个分数很低,在真实交易里大概率亏损。继续模拟?")
Output Template
Workflow 2 — Check Portfolio & Live P&L
Triggers: "我的仓位", "模拟盈亏", "portfolio", "how am I doing", "查看持仓"
Steps
Output Template
Workflow 3 — Simulate a Sell / Close Position
Triggers: "模拟卖出", "平仓", "paper sell", "close position", "我想卖掉X"
Steps
Exit Quality Score (add-on to Trade Score)
After closing, compute the Exit Score (0–20 bonus pts added to Trade Score):
| Condition | Points | |
|---|---|---|
| Sold at profit (P&L > 0) | +10 | |
| Exited while trend was still up (last 3 candles rising) | +5 | |
| Exited before momentum reversed (kline shows peak was just before) | +5 | |
| Sold at loss but cut quickly (held < 48h, loss < 10%) | +3 (damage control credit) | |
| Panic sold at bottom (price recovered >20% after exit) | −10 |
Output Template
Workflow 4 — Trading Report & Ability Score
Triggers: "我的交易能力", "总结报告", "评分", "trading report", "score my performance", "am I ready to trade for real"
Steps
Aggregate all closed trades in the session:
Also fetch one signal snapshot for context:
Use this to check: how many of the user's winning trades had smart money alignment?
Ability Verdict
| Avg Score | Verdict | |
|---|---|---|
| ≥ 80 | 🏆 优秀交易者 — 你的决策非常有纪律。考虑用小额真实资金开始练习,先从 $50–$100 起步。 | |
| 60–79 | 🟢 进阶中 — 基本面分析不错,继续练习出场时机。还不建议用真实资金。 | |
| 40–59 | 🟡 新手阶段 — 有闪光点,但几个关键维度需要加强(见下方)。继续模拟。 | |
| < 40 | 🔴 需要更多练习 — 目前的交易模式在真实市场中风险很高。请先多做模拟交易。 |
Output Template
Conversational Rules
Session State
- Always track: cash balance, open positions (entry price, qty, chain, address), closed trades, scores.
- Opening balance is always $10,000 USDC (virtual).
- On every session start: read
~/.web3-trade-simulator/portfolio.json. If found, resume from saved state and tell the user: "已加载你的历史仓位,继续上次的训练。" If not found, create it with the default $10,000 balance. - After every trade action: write the full updated state back to
~/.web3-trade-simulator/portfolio.jsonimmediately — do not wait until the end of the conversation. - If user wants to reset: overwrite the file with a fresh $10,000 state and confirm: "虚拟仓位已重置,重新从 $10,000 开始。"
- Remind user at the start: "这是虚拟资金,不会动你的真实钱包。"
Language
- Match the user's language (Chinese / English). Default to Chinese for mainland users.
- In Chinese: 大白话. Explain jargon inline: "止损(就是设定一个价格,跌到那就卖掉,防止继续亏)".
Safety Gate (Same as web3-safe-guide)
- Honeypot → Block, even in simulation.
- Illiquid token (liquidity < $10K) → Warn: "流动性太低,模拟交易意义不大,真实情况下你可能根本卖不出去。"
Educational Moments (inject once per session)
- First simulated buy → explain what "price impact" means.
- First time user picks a token with smart money signal → explain what smart money is.
- First loss → "亏损是学习的一部分,关键是搞清楚为什么亏。"
- First win → "先别急着觉得自己是天才,看看决策评分,赢了可能只是运气。"
End of Every Reply
Error Handling
| Situation | Response | |
|---|---|---|
| Token not found | "没找到这个代币,换个名字或者给我合约地址试试?" | |
| Honeypot in simulation | "☠️ 骗局代币。换一个真正值得研究的标的。" | |
| Insufficient virtual cash | "虚拟余额不够了(剩 $XX),要不要重置回 $10,000 重新开始?" | |
| User asks to trade with real money | "随时可以!先用 web3-safe-guide 做一个完整的安全评估,我帮你走流程。" | |
| CLI error | "数据获取失败,稍后再试。我先用上次价格估算给你参考。" |
onchainos CLI Commands Used
| Sub-skill | Commands | |
|---|---|---|
okx-dex-token | token search, token price-info, token holders | |
okx-dex-market | market kline, market signal-list | |
okx-dex-swap | swap quote (read-only — never swap swap or swap approve) |
This skill NEVER callsswap swaporswap approve. It is read-only by design.
Example Conversations
User: "帮我模拟买 100 USDC 的 BONK" → Workflow 1: token search → safety check → swap quote (entry price) → kline + signal check → score entry → record in portfolio → show decision score breakdown in Chinese
User: "我的仓位现在怎么样" → Workflow 2: price-info for each open position → compute P&L → kline momentum → show updated portfolio with live values in Chinese
User: "模拟卖掉 BONK" → Workflow 3: swap quote (exit price) → compute P&L + duration → exit score → full trade debrief in Chinese
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 96 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/web3-trade-simulator/SKILL.md
Popularity
- 7-day trend score
- 0
Refreshed 18 Mar 2026, 7 months ago.
Extension points
pluginCustom plugin integration
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/wy51ai/web3-starter-kit/web3-trade-simulator)