optimize · marketcalls — Claude Code skill: what it does, how to install
Generates Python parameter grid search scripts for VectorBT strategies, plots Plotly heatmaps, and benchmarks returns against NIFTY.
npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a claude-codeWhat optimize does
The skill creates an optimization script named {symbol}_{strategy}_optimize.py inside backtesting/{strategy_name}/. The generated code retrieves market data using OpenAlgo API or an optional read-only DuckDB database. By default, it computes technical indicators using openalgo.ta and cleans entry signals with ta.exrem(), avoiding VectorBT built-in indicators unless TA-Lib is explicitly requested.
Optimization runs in a tqdm loop across strategy parameter grids. It tracks total return, Sharpe ratio, max drawdown, and trade count for each run. Alongside printing top-10 parameter sets, the script outputs two dark Plotly heatmaps, saves raw results to CSV, and generates a comparison table against the NIFTY index.
Because the generated code embeds Indian market delivery fees (fees=0.00111, fixed_fees=20) and specific lot sizes for NIFTY and BANKNIFTY futures, running backtests for other exchanges requires manually editing commission logic and lot configurations.
What it helps with
- run parameter grid searches for VectorBT trading strategies
- generate Plotly heatmaps for total return and Sharpe ratio
- calculate trade performance with Indian equity delivery fees
- compare custom strategy metrics against the NIFTY benchmark
Example prompts for the agent
- /optimize ema-crossover RELIANCE NSE D
- /optimize rsi SBIN
- /optimize supertrend NIFTY NFO 5m
How to install optimize
1Skills CLI (Vercel)
npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a claude-code2Manually: put the files in place
~/.claude/skills/optimize/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/optimize/
1Skills CLI (Vercel)
npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a codex2Manually: put the files in place
.agents/skills/optimize/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a cursor2Manually: put the files in place
.cursor/skills/optimize/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install marketcalls/vectorbt-backtesting-skills optimize2Manually: put the files in place
.github/skills/optimize/SKILL.md1Skills CLI (Vercel)
npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a gemini-cli2Platform CLI
gemini skills install https://github.com/marketcalls/vectorbt-backtesting-skills --path .claude/skills/optimizeManually: put the files in place
.devin/skills/optimize/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about optimize
How to install optimize in Claude Code?
Run npx skills add marketcalls/vectorbt-backtesting-skills --skill optimize -a claude-code in your terminal.
What does the optimize skill do?
It writes a standalone Python script that performs grid optimization on VectorBT strategies, visualizes metrics via Plotly heatmaps, and saves results to CSV.
What data sources does the generated script support?
It connects to the OpenAlgo historical data API using .env credentials, or queries local market data files directly via DuckDB.
Similar skills
Everything for Finance and business →optimize by other authors comes first.
