Skip to content
SkillMixed2,284 tokens in use

Backtest Expert

A backtesting method that tries to break a strategy, plus a script that scores backtest statistics across five dimensions and writes a JSON and Markdown report.

7,426 stars on the repository MIT Licensetradermonty/claude-trading-skills · skills/backtest-expert

Skill label

measured from its files

What installing Backtest Expert costs

85tokens idle
2,284tokens in use
5,881tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Mixed

Script-ledMixedModel-led

Scripts do part of the work and the model's judgement does the rest.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/evaluate_backtest.py.

Touches

shelldetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter12/15
  • Structure25/25
  • Examples9/15
  • Output10/15
  • Size15/15
  • Resources7/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 120
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    scripts/evaluate_backtest.py, line 580

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance52
  • Community68
  • Maturity36
  • Momentum40

Backtest Expert gives the agent a step-by-step method for testing systematic trading strategies: state the hypothesis, codify the rules, run the test over several years, stress the parameters and costs, then check out-of-sample results.

It also ships a Python script that takes the summary numbers of a backtest (trades, win rate, average win and loss, drawdown, years tested, parameter count, whether slippage was tested) and scores them. The script reports a decision, the reasons that blocked it, and red flags. It is a review aid for research and does not give investment advice.

What You Get

  • A six-step backtesting workflow from hypothesis to out-of-sample validation
  • A Python script that scores sample size, expectancy, risk management, robustness and execution realism
  • A JSON report and a Markdown report with the decision, blocking reasons, red flags and input values
  • Two reference files, one on methodology and one on common failure patterns

Common Use Cases

  • Checking whether a backtest result is believable before relying on it
  • Finding which assumptions (slippage, parameter values, test period) a strategy depends on
  • Setting a personal drawdown limit and seeing whether a backtest exceeds it
  • Reviewing a strategy for signs of look-ahead bias or curve fitting

What Makes It Unique

The script treats its decision field as the authority and the 0-100 score as a diagnostic. Gates such as fewer than 30 trades, a drawdown of 50 percent or more, no slippage test, a test period under five years or results that look too good override a high score. Expectancy and profit factor are computed in exact decimal arithmetic so a break-even strategy cannot pass on floating point noise.

Under The Hood

The agent reads SKILL.md and, when needed, two Markdown reference files. To score a backtest it runs one local Python script with command line numbers; the script uses only the standard library, makes no network calls and needs no API keys. It writes a timestamped JSON file and a Markdown file into an output folder (reports/ by default), so it creates files on disk. Nothing else is fetched or executed.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers