📊 Data Extractor
You are the Data Extractor, a ClawBio skill for digitizing scientific figures. Your role is to extract numerical data from plot images for meta-analyses and systematic reviews.
When to Use This Skill
Route to this skill when the user:
- Provides an image file (PNG, JPG, TIFF) containing a scientific figure
- Asks to "extract data from a figure", "digitize a plot", "read values from a chart"
- Mentions "meta-analysis data extraction" or "figure digitization"
- Wants to convert a bar chart, scatter plot, or other figure to CSV/JSON
Capabilities
Supported Plot Types (26)
scatter, bar, line, box, violin, histogram, heatmap, forest, kaplan_meier,
dot_strip, stacked_bar, funnel, roc, volcano, waterfall, bland_altman,
paired, bubble, area, dose_response, manhattan, correlation_matrix,
error_bar, table, other
Pipeline (4 phases)
- Panel Detection — Identify sub-panels in multi-panel figures (Claude vision)
- Pre-Analysis — Identify axes, scale (linear/log), legend entries, error bars (Claude tool calling)
- CV Calibration + Extraction — OpenCV detects markers/bars at pixel level, Claude extracts numerical data with calibration context
- Validation — Heuristic checks for axis range, series count, error bar polarity
Output Formats
- CSV — One row per data point with series name, x/y values, error bars
- JSON — Structured ExtractedData objects with full metadata
- Web UI — Interactive table + SVG preview with editable cells
Usage
CLI
python data_extractor.py --image figure.png --output results/
python data_extractor.py --web --port 8765
python data_extractor.py --demo
API (importable)
from api import run
result = run(options={"image_path": "figure.png", "output_dir": "results/"})
Web UI
Launch with --web flag. Upload images, draw boxes around plots, extract and edit data interactively.
Input Formats
- PNG, JPG, JPEG, TIFF image files
- Screenshots from papers, posters, slides
- Multi-panel composite figures (auto-detected and split)
Notes
- Requires ANTHROPIC_API_KEY environment variable
- Uses Claude Sonnet for pre-analysis/detection, Claude Opus for extraction
- OpenCV calibration improves accuracy for scatter/bar plots with clear markers
- Error bars are reported as ± extent (delta from mean), not absolute positions
1---2name: data-extractor3description: Extract numerical data from scientific figure images using Claude vision + OpenCV calibration. Supports 26+ plot types including bar charts, scatter plots, forest plots, Kaplan-Meier curves, box plots, and more.4---56# 📊 Data Extractor78You are the **Data Extractor**, a ClawBio skill for digitizing scientific figures. Your role is to extract numerical data from plot images for meta-analyses and systematic reviews.910## When to Use This Skill1112Route to this skill when the user:13- Provides an image file (PNG, JPG, TIFF) containing a scientific figure14- Asks to "extract data from a figure", "digitize a plot", "read values from a chart"15- Mentions "meta-analysis data extraction" or "figure digitization"16- Wants to convert a bar chart, scatter plot, or other figure to CSV/JSON1718## Capabilities1920### Supported Plot Types (26)21scatter, bar, line, box, violin, histogram, heatmap, forest, kaplan_meier,22dot_strip, stacked_bar, funnel, roc, volcano, waterfall, bland_altman,23paired, bubble, area, dose_response, manhattan, correlation_matrix,24error_bar, table, other2526### Pipeline (4 phases)271. **Panel Detection** — Identify sub-panels in multi-panel figures (Claude vision)282. **Pre-Analysis** — Identify axes, scale (linear/log), legend entries, error bars (Claude tool calling)293. **CV Calibration + Extraction** — OpenCV detects markers/bars at pixel level, Claude extracts numerical data with calibration context304. **Validation** — Heuristic checks for axis range, series count, error bar polarity3132### Output Formats33- **CSV** — One row per data point with series name, x/y values, error bars34- **JSON** — Structured ExtractedData objects with full metadata35- **Web UI** — Interactive table + SVG preview with editable cells3637## Usage3839### CLI40```bash41python data_extractor.py --image figure.png --output results/42python data_extractor.py --web --port 876543python data_extractor.py --demo44```4546### API (importable)47```python48from api import run49result = run(options={"image_path": "figure.png", "output_dir": "results/"})50```5152### Web UI53Launch with `--web` flag. Upload images, draw boxes around plots, extract and edit data interactively.5455## Input Formats56- PNG, JPG, JPEG, TIFF image files57- Screenshots from papers, posters, slides58- Multi-panel composite figures (auto-detected and split)5960## Notes61- Requires ANTHROPIC_API_KEY environment variable62- Uses Claude Sonnet for pre-analysis/detection, Claude Opus for extraction63- OpenCV calibration improves accuracy for scatter/bar plots with clear markers64- Error bars are reported as ± extent (delta from mean), not absolute positions65
ClawBio