技能 backtest-expert

📦

backtest-expert

Name: backtest-expert
Author: tradermonty

內容修訂 r1 安全 ⚙️ 外部命令

Evaluate Trading Backtests for Robustness

Backtest results can hide overfitting, weak samples, and unrealistic execution assumptions. This skill applies structured scoring, stress tests, and bias checks.

支援: Claude Codex Code(CC)

🥈 80 銀級

Maintain this Skill? Claim it 🥈 80 銀級

使用我的智慧代理安裝

將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。

智能體請求

Review the Skillstore skill "backtest-expert" from https://skillstore.io/skills/tradermonty-backtest-expert.md and its manifest at https://skillstore.io/api/skills/tradermonty-backtest-expert/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。

Agent 可讀資源

當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。

Markdown 詳情 GET /skills/tradermonty-backtest-expert.md 已簽名 manifest GET /api/skills/tradermonty-backtest-expert/manifest 已簽名 lockfile GET /api/skills/tradermonty-backtest-expert/lockfile

測試它

正在使用「backtest-expert」。 Evaluate 150 trades, 62 percent wins, 1.8 percent average win, 1.2 percent average loss, 15 percent drawdown, eight years, three parameters, with slippage tested.

預期結果:

Deploy, 81 out of 100. The sample, expectancy, drawdown, test duration, parameter count, and execution assumptions support further consideration. Independent validation remains necessary.

正在使用「backtest-expert」。 Review 24 trades over two years with eight parameters and no slippage test.

預期結果:

Verdict: Abandon.
The sample is below the minimum threshold.
The test period may omit important market regimes.
Eight parameters indicate substantial overfitting risk.
Execution costs have not been validated.

正在使用「backtest-expert」。 Explain why a profitable strategy fails after realistic trading costs.

預期結果:

The reported edge is smaller than expected slippage and commissions. Test conservative fills and nearby parameters before deciding whether the hypothesis remains viable.

安全審計

安全

v1 • 2026/9/27 開啟版本化報告

All 17 static findings are false positives caused by trading prose, Python test identifiers, Markdown code fences, and inline code formatting. The skill performs local calculations and writes requested reports, with no network access, secret handling, prompt injection, or dynamic shell execution found.

已掃描檔案

1,839

分析行數

複核項目

已忽略的誤報

風險因素

⚙️ 外部命令 (6)

SKILL.md:119-130 SKILL.md:130-181 SKILL.md:181-185 SKILL.md:185-186 SKILL.md:186-191 SKILL.md:191-204

最近一次完成的靜態與語意稽核未偵測到已確認的安全發現。這並不能證明該 Skill 沒有副作用。

審計單位: codex

分享與引用此報告

分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。

開啟版本化報告

複製報告連結

https://skillstore.io/skills/tradermonty-backtest-expert/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/tradermonty-backtest-expert/security.svg)](https://skillstore.io/skills/tradermonty-backtest-expert?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/tradermonty-backtest-expert?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/tradermonty-backtest-expert/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/tradermonty-backtest-expert.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

學術引用 (APA · BibTeX · CFF)

APA 引用

tradermonty. (2026). backtest-expert security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/tradermonty-backtest-expert/audits/1

BibTeX 引用

@techreport{tradermonty-tradermonty-backtest-expert-2026,
  author = {tradermonty},
  title = {backtest-expert security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/tradermonty-backtest-expert/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "backtest-expert security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "tradermonty"
date-released: "2026-09-27"
url: "https://skillstore.io/skills/tradermonty-backtest-expert/audits/1"
identifiers:
  - type: other
    value: "skillstore:tradermonty-backtest-expert:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 評分

評分依據證據置信度: 中

架構

可維護性

內容

社群

規範符合度

你可以打造什麼

Screen a New Strategy

Review early backtest metrics and identify the highest-priority validation gaps before further development.

Challenge Research Results

Apply bias checks, parameter stress tests, and walk-forward criteria during a formal research review.

Teach Robust Validation

Demonstrate why sample size, execution friction, market regimes, and out-of-sample testing affect confidence.

試試這些提示

Review Basic Metrics

Evaluate this backtest: [trades], [win rate], [average win], [average loss], [drawdown], [years], [parameters], and [slippage tested]. Explain the verdict simply.

Build a Stress Test Plan

Create a stress test plan for this strategy: [rules]. Cover parameter ranges, slippage, commissions, fill quality, sample size, and market regimes.

Diagnose a Failed Backtest

Analyze these failed backtest results: [results]. Identify breaking points, likely biases, weak assumptions, and the next tests that could reject or refine the hypothesis.

Conduct an Adversarial Validation

Audit this strategy and evidence: [strategy], [metrics], [periods], [cost assumptions], [parameter tests], and [out-of-sample results]. Produce a prioritized robustness assessment and deployment conditions.

最佳實務

Define entry, exit, sizing, filters, and the trading universe before testing.
Use multiple market regimes, out-of-sample periods, and conservative execution costs.
Prefer broad performance plateaus over isolated optimal parameter values.