Shadow Trading Dashboard

Manage and update Steven's A-share shadow trading dashboard including positions, trades, metrics, equity curve, and trade history.

Security Scan

This is an instruction-only skill that coherently documents updating a local 'trade/' dashboard (read/write local files); the functionality aligns with its description, but the source is unknown and the metadata omits declared config paths — review and back up files before use.

Purpose & Capability

The skill's name and description match the actions in SKILL.md: managing a local A-share shadow trading dashboard. It requires reading and writing files under a trade/ workspace which is consistent with the purpose. Minor inconsistency: registry metadata lists no required config paths, but SKILL.md expects files (e.g., ~/.openclaw/workspace/trade/...).

Instruction Scope

SKILL.md explicitly instructs the agent to read and write specific local files (positions.json, orders.jsonl, CSVs, metrics, equity curve, index.html) and to open the dashboard in a browser. All referenced file operations are within the stated trade/ dashboard scope and no external network endpoints or unrelated system files are referenced. It does assume filesystem access to ~/.openclaw/workspace/trade/ (or a similar workspace) which the metadata did not declare.

Install Mechanism

No install spec and no code files — instruction-only. This has low installation risk because nothing is downloaded or written by an installer.

Credentials

The skill does not request environment variables, credentials, or external service access. Its file read/write needs are proportional to the dashboard-editing purpose.

Persistence & Privilege

The skill is not marked always:true and does not request persistent modification of other skills or system-wide config. Autonomous invocation is allowed (platform default) but not by itself a concern here.

Assessment

This skill appears to simply instruct edits to local files in trade/. Before installing: (1) confirm where your trade/ workspace actually resides (SKILL.md references ~/.openclaw/workspace/trade/), (2) back up the trade/ folder and index.html so you can recover if something is overwritten, (3) review the actual files (positions.json, orders.jsonl, trades.csv, etc.) to ensure they contain no secrets you don't want an agent to modify or expose, (4) if you run the agent with file access, limit its permissions or run in a sandbox/test workspace first, and (5) note the skill's source/homepage is unknown — if provenance matters, consider reaching out to the author or using a locally curated copy of the SKILL.md to avoid unexpected changes.

Dashboard Location

trade/dashboard/index.html — 影子盘看板主文件，双击用浏览器打开

Core Data Files

File	Purpose
trade/dashboard/index.html	Dashboard HTML
trade/dashboard_data/metrics.json	KPI metrics
trade/dashboard_data/equity_curve.json	Equity curve
trade/data/positions.json	Current positions
trade/data/orders.jsonl	Order log
trade/trades.csv	Open positions CSV
trade/closed_positions.csv	Closed positions CSV
trade/trade_history.md	Human-readable trade history

Data Schema

positions.json

{
  "symbol": "300274",
  "name": "阳光电源",
  "quantity": 300,
  "cost": 163.16,
  "current_price": 166.59,
  "stop_loss": 160.00,
  "tags": ["储能","龙头"],
  "entry_reason": "储能龙头ROE29%，回调支撑买入"
}

metrics.json

{
  "current_equity": 1001029.00,
  "position_market_value": 49977.00,
  "cash": 951037.32,
  "unrealized_pnl": 1029.00,
  "realized_pnl": 0.0,
  "total_return_pct": 0.103,
  "trade_count": 1,
  "win_count": 0,
  "loss_count": 0,
  "max_drawdown": 0.0
}

Standard Workflows

Opening the Dashboard

open ~/.openclaw/workspace/trade/dashboard/index.html

After Any Trade (Buy/Sell/Stop)

1. Update trade/data/positions.json
2. Append to trade/data/orders.jsonl
3. Update trade/trades.csv and trade/closed_positions.csv
4. Update trade/dashboard_data/metrics.json
5. Update trade/dashboard_data/equity_curve.json
6. Rebuild trade/dashboard/index.html (update DATA object + equity curve)
7. Update trade/trade_history.md
8. Run open trade/dashboard/index.html

Building the Dashboard HTML

The dashboard is a self-contained HTML with an embedded DATA object. Key fields to update:

• current_equity, cash, position_value, unrealized_pnl, realized_pnl
• positions[] array (name, symbol, quantity, cost, current_price, stop_loss, tags)
• trade_history[] array (date, time, symbol, name, action, direction, price, quantity, amount, fee, pnl, tags)
• closed_positions[] for closed trades
• equity_curve[] for the curve

Fees

• A股模拟手续费：0.03% 单边（买卖各收一次）
• Fee = price × quantity × 0.0003

Equity Calculation

• current_equity = cash + Σ(position market value)
• Market value = current_price × quantity

Dashboard Update Checklist

After any trade, ALL of these must be updated:

• trade/data/positions.json
• trade/data/orders.jsonl (append new order)
• trade/trades.csv (open positions)
• trade/closed_positions.csv (closed positions)
• trade/dashboard_data/metrics.json
• trade/dashboard_data/equity_curve.json
• trade/dashboard/index.html (DATA object + equity_curve)
• trade/trade_history.md