← Back to list
backtrader Checked
Event-driven backtesting with bar-by-bar execution, complex order types, multiple analyzers, and …
8.6Overall rating
Installation
🤖 Install via AI
Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.
Install the "backtrader" skill by following the instructions at https://skill123.me/install/backtrader.
⌨️ Command line install
Run in your terminal — downloads and installs to ~/.claude/skills/.
curl -fsSL https://skill123.me/install/backtrader.sh | bash
📦 Download ZIP
Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.
⬇ Download v1.0.0 · 20 KBAbout this skill
Overview
Event-driven backtesting with bar-by-bar execution, complex order types, multiple analyzers, and custom indicators
Source
- Repo: https://github.com/agiprolabs/claude-trading-skills
- Path: skills/backtrader
Score breakdown
Trigger
6.0
What-only description with framework keywords; strong Use-backtrader-when / Use-vectorbt-when guidance in body but not in description; no trigger variants, no Chinese keywords.
Structure
9.3
348-line SKILL.md; api_guide + strategy_patterns references and two scripts; Files table; inline See-references pointers.
Workflow
8.0
Task-oriented guide with runnable EMA-crossover and bracket-order demo scripts (synthetic data, seeded), realistic-commission and slippage configuration, lookahead avoidance via set_coo; moderate failure-path coverage (order status notifications documented).
Content
10.0
Concise tables, complete code snippets with parameters, consistent formatting.
Engineering
10.0
Valid frontmatter; shebangs, deps declared in Quick Start, --demo flags default to synthetic data; all referenced paths exist.
Security
10.0
Manifest: no network, no file/env/credential reads (plot example writes a PNG to cwd, benign); pure local backtesting on synthetic data; capabilities match declared scope. No injection, no secrets.
Clear framework guide: event-driven vs vectorized decision table (use backtrader vs vectorbt), core object walkthrough, order types incl. brackets, crypto-specific commission/slippage/lookahead guidance (cheat-on-open), two demo scripts on synthetic data. Description is What-only; When/When-Not lives in body only.