Precise T Trading
ReviewAudited by ClawScan on May 10, 2026.
Overview
Review before installing: the skill does not show broker order placement or credential theft, but it gives real-money trading recommendations while relying on hard-coded or simulated data and has a few under-disclosed local execution details.
Use this skill cautiously and preferably for education or paper trading. Verify all market data and replace the mock/hard-coded assumptions before considering real trades. Do not run it as administrator, review the hard-coded log path, and remember that the artifacts do not show actual broker order execution or credential handling.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user could rely on seemingly precise buy/sell and position-size recommendations that are partly based on placeholder data, potentially causing financial losses.
The source shows hard-coded trade history and simulated historical prices, while the skill presents itself as a professional T+0 trading decision system. This mismatch can make the output appear more evidence-backed than it is.
T_HISTORY = { 'total_trades': 20, 'wins': 13, ... } ... "获取历史数据（模拟，实际可接入tushare）" ... np.random.normal(0, 0.03)Treat this as educational or paper-trading software unless the data inputs are replaced with verified real trading history and the output clearly states its assumptions and limitations.
The script may create files outside the expected skill directory or fail on systems without that Windows path.
The monitor creates and writes to a hard-coded absolute path, while SKILL.md says logs are written under ./monitor_logs/. This is purpose-aligned logging, but the actual path is under-disclosed and not user-configurable.
log_dir = r"I:\OpenClawWorkspace\stocks\监控日志" ... os.makedirs(log_dir, exist_ok=True)
Change the log directory to a user-chosen or relative path before running the monitor.
Opening the dashboard contacts the quote provider repeatedly and depends on that provider’s script response being safe.
The dashboard uses JSONP-style dynamic script loading from Tencent Finance and refreshes every 30 seconds. This is aligned with the real-time quote purpose, but it means the local dashboard executes remote script responses.
script.src = `https://qt.gtimg.cn/q=${code}&callback=${callbackName}`; ... setInterval(refreshData, 30000);Only open the dashboard if you are comfortable with the Tencent Finance request, and prefer a safer data-fetching design if modifying the skill.
Running as administrator increases the impact of any bug or unexpected behavior in the scripts.
The troubleshooting guidance suggests elevated privileges, although the visible scripts only need network access and local log/file writes. This is not required by the declared purpose.
"Permission denied" | Run as administrator on Windows
Avoid running this skill with administrator privileges; instead fix file permissions or choose a writable log/cache directory.