Skip to content
Get packs

Backtesting Frameworks

Guides the agent to build trading strategy backtests that avoid look-ahead and survivorship bias.

Source Download skill
Repository
wshobson/agents
Stars
42,095
Licence
MIT
Files
2 files

What it does

This skill teaches the agent to design and review historical tests of trading strategies, the kind that check how an idea would have performed on past market data. Once installed, the agent follows safer habits: using only information that was available at each point in time, splitting data into training, validation and test sets, using walk-forward analysis, and including trading costs. It is guidance and structure rather than a ready-made trading tool.

Great for

  • Quant developers building or reviewing a backtesting pipeline
  • Traders or analysts who want to check whether a strategy's results hold up out of sample
  • Students learning why backtests often overstate performance

Skip it if

  • Giving investment advice or deciding trades: it describes testing methods, not whether any strategy will make money
  • Live trading execution: the skill covers historical simulation only and does not connect to brokers or markets
  • Anyone expecting a finished backtesting library: the files describe principles and patterns; the referenced details file was not reviewed here for code specifics

Try asking

  • “Help me design a walk-forward backtest for a moving-average crossover strategy on daily stock data”
  • “Review my backtest code for look-ahead bias and survivorship bias”
  • “Set up training, validation and test splits for a mean-reversion strategy”
  • “Add realistic transaction cost and slippage assumptions to my backtest”
Step-by-step guide: how to use Backtesting FrameworksInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    backtesting-frameworks.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip backtesting-frameworks.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip backtesting-frameworks.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • State your data source and whether it is point-in-time, since the skill's bias advice depends on it
  • Ask the agent to write the code and run it on sample data before trusting any results
  • Treat any backtest output as an estimate; validate on data the strategy never saw

What's inside

  • references/details.md
  • SKILL.md

Redistributed unmodified from wshobson/agents under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.