sn-da-image-caption

Generate vision captions from images and parse them into tabular data.

2|Updated May 19, 2026
One-click install
npx skills add https://github.com/aiyinluya/SenseNova-Skills-Studio --skill sn-da-image-caption-aiyinluya
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sn-da-image-caption
Source: https://github.com/aiyinluya/SenseNova-Skills-Studio/tree/main/skills/sn-da-image-caption
Command: npx skills add https://github.com/aiyinluya/SenseNova-Skills-Studio --skill sn-da-image-caption-aiyinluya

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Pillow, openai, pandas, matplotlib, openpyxl, and includes scripts (resource) components.

What problem does it solve?

This Skill helps you understand image content and extract structured data from charts, tables, UI screenshots, and diagrams so you can analyze it and reuse it in spreadsheets.

Core Features & Use Cases

  • Vision-based image captioning: Converts supported image files into detailed text descriptions using a pre-configured vision script.
  • Data parsing and restructuring: Turns caption output (often Markdown tables) into structured data (e.g., a DataFrame) for downstream analysis.
  • Re-visualization and export: Regenerates visualizations and exports extracted results to Excel/CSV for reporting and sharing.
  • When to use: Best for extracting values from chart/table/UI/diagram screenshots and for transforming the extracted data into Excel/CSV or charts.

Quick Start

Upload a chart or table image and ask: “Analyze this image, extract all numeric data into a Markdown table, then export the results to Excel.”

Frequently Asked Questions about sn-da-image-caption

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

FAQPage Schema
How do I extract data from a chart or table image into Excel?

To extract data from a chart or table image into Excel, upload the image and request numeric extraction. The Skill generates a vision caption, parses it into structured data, and exports the results directly to Excel or CSV format.

Can I use a vision model to parse UI screenshots into structured data?

Yes, you can use a vision model to parse UI screenshots into structured data. The Skill converts image files into detailed text descriptions using type-aware prompting, then restructures the output into a DataFrame for analysis.

What is the best way to convert image-based tables to CSV exports?

The best way to convert image-based tables to CSV exports is through a vision-based image captioning pipeline. It captures the table structure in Markdown, parses it into a DataFrame, and writes the extracted values to a CSV file.

Does this image captioning approach work with matplotlib and pandas for re-visualization?

Yes, this image captioning approach works with matplotlib and pandas for re-visualization. After extracting structured data from the image, the pipeline uses pandas to manage the DataFrame and matplotlib to regenerate visualizations for reporting.

How do I extract numeric values from a diagram screenshot using OCR from screenshots?

To extract numeric values from a diagram screenshot, the Skill applies a vision model to generate a detailed text caption, identifies the numeric data within the output, and parses those values into a structured DataFrame for downstream analysis.

What are the limitations of using a vision model for complex chart data extraction?

Limitations of using a vision model for complex chart data extraction include potential misinterpretation of overlapping data points or unclear labels in low-resolution images, which can affect the accuracy of the parsed DataFrame and subsequent Excel export.