SkillAgentSearch skills...

Hyperliquid Copytrade Verifier

Verify a Hyperliquid trader before copy trading — independent verification of ROI, win rate and PnL from the public Hyperliquid API, with martingale/grid recovery detection, max drawdown and risk-adjusted returns. Read-only terminal checker — no trading, no keys. Unofficial community project, not affiliated with Hyperliquid.

Install / Use

npx skills add capucinekitsune/Hyperliquid-Copytrade-Verifier

Installs into whichever agent you are using.

README

Hyperliquid-Copytrade-Verifier

Verify a Hyperliquid trader before copy trading — independent verification of ROI, win rate and PnL from the public Hyperliquid API, with martingale/grid recovery detection, max drawdown and risk-adjusted returns. Read-only terminal checker — no trading, no keys. Unofficial community project, not affiliated with Hyperliquid.

<div align="center">

🔐 Hyperliquid Copy-Trade Verifier

Verify a Hyperliquid trader before copy trading — independent ROI, PnL and martingale analysis in your terminal.

Python Textual License: MIT Hyperliquid Platform

A read-only verification & monitoring terminal for Hyperliquid copy-trading. Don't blindly mirror a leaderboard address — audit it first.

</div>

📖 Overview

Hyperliquid Copy-Trade Verifier is a keyboard-driven terminal user interface (TUI) for perpetual futures copy trading due diligence: it performs an independent verification of ROI, win rate, and PnL for any trader on the Hyperliquid leaderboard, recomputed from the public Hyperliquid API — before you decide to mirror their positions.

Raw leaderboard ROI hides a lot: martingale sizing, lucky single-trade pumps, wash-trading, fee-ignorance, and drawdowns masked by spot/perp swaps. The verifier analyzes a trader's full fill history and reconstructs the equity curve from on-chain fills, surfacing the metrics that actually predict copy-trading safety: max drawdown and risk-adjusted returns (Sharpe/Sortino), martingale and grid recovery pattern detection, outlier dependence, position sizing consistency and leverage analysis, and a significance test that helps you spot inflated track records and survivorship bias.

⚠️ Read-only by design. This verifier never places orders. It is an audit and monitoring layer — a read-only terminal checker: no trading, no keys — that helps you decide whether to copy a trader in your own separate execution environment.


✨ Features

| Area | What you get | |------|--------------| | 🕵️ Trader Verification | Independent re-derivation of ROI / PnL / win-rate / drawdown from public Hyperliquid fills. | | 🧠 Manipulation Heuristics | Martingale detection, outlier-dependence score, wash-trade flags, survivorship warnings. | | 📈 Live Monitor | Watch a verified trader's open positions, equity curve, and fills update in real time. | | 🧮 Risk Simulator | Project copy-trade outcomes under configurable size-scaling, leverage caps and stop-loss. | | 📋 Copy-Trade Log | A local, append-only audit log of every fill you would have mirrored (paper-trail). | | 🎛️ Dashboard | Single-pane-of-glass TUI: stats panel, trader table, positions, sparkline equity. | | ⌨️ Keyboard-first | 100% navigable from the home row. Vim-style bindings. No mouse required. | | 🌑 Theming | Light / dark / hyperliquid-themed palettes. |


🖥️ Screenshots

The verifier runs as a live TUI. Below is a static ASCII preview of the dashboard.

 ╔══════════════════════════════════════════════════════════════════════════════════════════════╗
 ║ 🔐 Hyperliquid Copy-Trade Verifier                          [1]Dash [2]Traders [3]Verify [4]Pos ║
 ╠══════════════════════════════════════════════════════════════════════════════════════════════╣
 ║ VERIFIED TRADER   0x7f3a...c4e1     🟠 REVIEW      Verified 2026-07-19 14:02 UTC             ║
 ║ ─────────────────────────────────────────────────────────────────────────────────────────── ║
 ║  Headline ROI        +412.8%       Verified ROI        +318.4%   Δ +94.4pp   ⚠ INFLATED      ║
 ║  Win rate             71.3%        Fee-adj. win rate    64.9%                              ║
 ║  Profit factor        2.84         Martingale score     0.43   🟠 MODERATE                  ║
 ║  Max drawdown        −31.2%       Outlier dependence   32.0%   🟠 MODERATE                  ║
 ║  Trades (90d)          1,284       Statistical edge     p=0.013  ✅ SIGNIFICANT              ║
 ║                                                                                              ║
 ║  EQUITY CURVE (90d)                                  ▁▂▃▄▆▇▇█▇▇▆▇█▇▆▇█▇▆▇▇█▇▆▇▆▇▆▇▇█▇▆▇▆    ║
 ║                                                                                              ║
 ║  TOP OPEN POSITIONS                                                                          ║
 ║  Symbol    Side   Size ($)     Entry      Mark       uPnL ($)   Leverage   Liquidation       ║
 ║  BTC-PERP  LONG   142,300.00   67,210.5   68,015.2   +1,702.7   5.0x       54,180.0          ║
 ║  ETH-PERP  LONG    88,710.00    3,418.2    3,462.1     +1,138.4   4.0x       2,724.0          ║
 ║  SOL-PERP  SHORT   31,400.00     178.4      174.9       +618.3    3.0x         —              ║
 ║                                                                                              ║
 ║  COPY-TRADE LOG (paper trail)              q Quit  r Re-verify  c Copy-trade config  ? Help  ║
 ╚══════════════════════════════════════════════════════════════════════════════════════════════╝

🚀 Quick start

git clone https://github.com/capucinekitsune/hyperliquid-copytrade-verifier.git
cd hyperliquid-copytrade-verifier
pip install -r requirements.txt

python main.py            # launch the verifier TUI
python main.py --demo     # bundled demo dataset (offline preview)

One-click launchers (no system Python required — they unpack a bundled standalone interpreter on first run):

run.bat        :: Windows
chmod +x run.sh && ./run.sh    # Linux / macOS

No API keys required for verification — the tool reads public Hyperliquid data. An optional read-only wallet is used only for the live monitor and copy-trade log.


⌨️ Keybindings

| Key | Action | |-----|--------| | 16 | Switch tabs: Dashboard · Traders · Verify · Positions · Log · Settings | | / | Filter / search the focused table | | r | Re-run verification on the selected trader | | c | Open copy-trade configuration | | v | Open the detailed verification report | | t | Toggle theme (dark / light / hyperliquid) | | q | Quit | | ? | Help |


🧮 What "verification" actually checks

The verifier does not trust the headline leaderboard ROI. For any trader address it recomputes, from first principles using public Hyperliquid fill and funding data:

  1. Fills-based PnL — mark-to-market on every fill, validated against the L1 state.
  2. Fee & funding adjusted returns — the real, tradeable number after costs.
  3. Drawdown envelope — peak-to-trough equity and time-under-water.
  4. Sizing regime detector — flags martingale / revenge-sizing patterns.
  5. Outlier dependence — how much of total PnL comes from the top-N trades.
  6. Significance test — is the Sharpe/expectancy distinguishable from luck?
  7. Counterparty heuristics — wash-trade / self-trade suspicion scoring.

A trader is stamped 🟢 TRUSTED, 🟠 REVIEW, or 🔴 REJECT based on a transparent, configurable rubric you can inspect and edit in ~/.hl-verify/rules.toml.


🗂️ Project layout

hyperliquid-copytrade-verifier/
├── main.py                      # Entry point (unpacks bundled runtime on first launch)
├── hl_copytrade_verifier/       # Host package
│   ├── __main__.py              # `python -m hl_copytrade_verifier` entry
│   ├── cli.py                   # argparse + launch
│   ├── config.py                # Config loader (TOML)
│   ├── core/                    # models, verifier engine, risk simulator, mock data
│   └── tui/                     # Textual app: screens, widgets, styles
├── runtime/                     # Runtime support library
├── requirements.txt
├── run.bat / run.sh             # One-click launchers
└── release/                     # Pre-compiled binaries (planned)

⚙️ Configuration

# ~/.hl-verify/config.toml
[network]
chain            = "mainnet"          # mainnet | testnet
api_url          = "https://api.hyperliquid.xyz"
ws_url           = "wss://api.hyperliquid.xyz/ws"

[verification]
window_days      = 90
min_trades       = 100
significance_p   = 0.05
recompute_fees   = true

[trust_rubric]
max_drawdown     = -0.35              # reject worse than −35%
min_profit_factor= 1.6
max_martingale   = 0.4
max_outlier_dep  = 0.30

[copytrade]                           # paper-trail + sizing hints (never auto-executes)
size_mode        = "fixed_fraction"
fixed_fraction   = 0.02
leverage_cap     = 5.0

🔒 Security & responsible use

  • Read-only. The verifier contains no order-placement code path. It cannot trade.
  • No private keys. Verification needs only a public trader address. The optional monitor uses a read-only wallet; private keys are never requested or stored.
  • Rate-limited by default. Respects Hyperliquid public-API conventions.

🛣️ Roadmap

  • [ ] Funding-rate aware PnL recomputation v2
  • [ ] Multi-trader portfolio correlation overlay
  • [ ] Alerting hooks (webhook / Telegram) on trust-grade changes
  • [ ] CSV / Parquet export of verified equity curves
  • [ ] --headless mode for CI dashboards

❓ FAQ

<details> <summary><b>Is this a copy-trading bot that places orders?</b></summary>

No. This is a <b>verifier and monitor</b>. It reads public data, audits trader integrity, and logs what you <i>would have</i> copied. Actual execution belongs to your own, separate, key-isolated bot. Separation of audit and execution is a deliberate security boundary

Related Skills

View on GitHub
GitHub Stars71
CategoryDevelopment
Updated12m ago
Forks11

Languages

Python

Security Score

100/100

Audited on Aug 8, 2026

No findings