chart-vision

Render, preprocess, detect patterns, and annotate chart images.

10|2|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/mahmoud20138/Tradecraft --skill chart-vision
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chart-vision
Source: https://github.com/mahmoud20138/Tradecraft/tree/main/plugins/tradecraft/skills/chart-vision
Command: npx skills add https://github.com/mahmoud20138/Tradecraft --skill chart-vision

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Chart data and visuals often live in images or screenshots. This skill provides an end-to-end pipeline to render OHLCV data into high-quality charts, preprocess screenshots to clean noise, detect candlestick and classical chart patterns, identify trendlines and S/R zones, and annotate results back onto the chart for reporting and analysis.

Core Features & Use Cases

  • Unified pipeline: render, preprocess, pattern-vision, trendline-S/R vision, pattern scanner, and annotation overlay.
  • Use Case: traders and analysts can generate annotated charts from screenshots or data to support decision making.
  • Use Case: researchers can extract and visualize chart patterns for datasets or publications.

Quick Start

Render an OHLCV chart from data or a screenshot, then preprocess, detect patterns and trendlines, and output an annotated chart image.

Frequently Asked Questions about chart-vision

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect candlestick patterns from a chart screenshot?

To detect candlestick patterns from a chart screenshot, you can preprocess the image to clean noise and then apply vision-based pattern detection. This pipeline identifies classical chart patterns and outputs an annotated chart image for analysis.

Can I render OHLCV data into a high-quality chart image?

Yes, you can render OHLCV data into a high-quality chart image using the mplfinance rendering stage. This transforms raw financial data into visual charts that can then be preprocessed, analyzed, and annotated.

What is the best way to identify trendlines and support resistance zones in an image?

The best way to identify trendlines and support resistance zones in an image is through a dedicated trendline and S/R vision analysis stage. This detects key levels directly from chart visuals and overlays them as annotations.

Does this chart image analysis pipeline work with OpenCV for preprocessing?

Yes, the chart image analysis pipeline works with OpenCV and scikit-image for preprocessing. These libraries are used to clean screenshot noise and prepare images for accurate pattern detection and annotation overlay.

How do I overlay annotations onto a chart for reporting?

You overlay annotations onto a chart for reporting by running detected patterns and trendline results through an annotation overlay stage. This generates a final annotated chart image that highlights candlestick patterns and S/R zones.

What are the limitations of using vision-based pattern detection on financial charts?

A limitation of vision-based pattern detection on financial charts is its reliance on image preprocessing quality; noisy screenshots can hinder accurate candlestick pattern identification and trendline analysis, requiring clean inputs for reliable annotated outputs.