Custom Indicator

This generates production-ready technical indicators with Numba JIT compilation, so you get O(n) performance on price data without writing C extensions. It scaffolds the full package: the core computation function decorated with @njit, a pandas-friendly public API, a Plotly charting script, and a benchmark suite. The workflow checks if your indicator already exists in openalgo.ta before creating anything, which saves you from reinventing RSI. Most useful when you need a custom signal like a Z-score oscillator or squeeze momentum detector and you want it fast enough to backtest on years of tick data without waiting. The Numba constraints are strict but documented: no fastmath, no pandas inside jitted code, explicit loops only.

skills.sh installs: 589
GitHub stars: 21
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add marketcalls/openalgo-indicator-skills --skill custom-indicator --agent claude-code

Installs into .claude/skills of the current project.

Files

Create a custom technical indicator by composing openalgo's Rust-core ta primitives with vectorized NumPy.

Arguments

$0 = indicator name (e.g., zscore, squeeze, vwap-bands, custom-rsi, mean-reversion). Required.

If no arguments, ask the user what indicator they want to build.

Instructions

Read the indicator-expert rules, especially:
- rules/custom-indicators.md — NumPy + ta-primitive patterns and templates
- rules/performance.md — Rust core performance, O(n) guarantees, benchmarking
- rules/indicator-catalog.md — Check if indicator already exists in openalgo.ta
Check first: If the indicator already exists in openalgo.ta (100+ indicators), tell the user and show the existing API
Create custom_indicators/{indicator_name}/ directory (on-demand)
Create {indicator_name}.py with:

File Structure

"""
{Indicator Name} — Custom Indicator
Description: {what it measures}
Category: {trend/momentum/volatility/volume/oscillator}
"""
import numpy as np
import pandas as pd
from openalgo import ta

SKILL.md on GitHub