Computer vision for bio-image preprocessing, feature detection, real-time microscopy. Color conversion, morphology, contour/blob detection, template matching, optical flow on fluorescence/brightfield. 10-100× faster than pure Python via C++. Use scikit-image for scientific morphometry/regionprops; OpenCV for real-time, video, classical feature extraction.
SkillScore91Maintained
Automatic
◆AI Engineer pack
Last verified 8/23/2026
What it changes
Passed 21 of 23 test cases with the skill, and 18 of 23 without it.
Without skill78%
With skill91%
Checks it passed that the model alone missed
Subtracts estimated background using cv2.subtract.
Calculates centroid coordinates using image moments cv2.moments.
Normalizes 16-bit images using cv2.normalize with cv2.NORM_MINMAX.
and 3 more in the full report
3 fixed18 passed either way2 still failing0 broke
Pass-rate lift
+13 pp
Tokens
+212%
Agent turns
0%
Response time
-18%
Model gemini-3.6-flashMeasured 8/23/2026v121 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 3
Prompt
I need a Python function that reads a 16-bit DAPI fluorescence microscopy image, normalizes it to 8-bit, enhances local contrast, performs automated binary segmentation, and measures nuclei morphology. The function should accept `image_path` and `min_area` parameters, filter out objects smaller than `min_area`, and return a pandas DataFrame containing columns for `area`, `cx`, `cy`, and `perimeter` for each detected nucleus.
What was checked, without and with the skill
Checked
Without skill
With skill
Uses cv2.imread with cv2.IMREAD_UNCHANGED flag when loading the image.
Missed
Passed
Normalizes 16-bit images using cv2.normalize with cv2.NORM_MINMAX.
Missed
Passed
Calculates centroid coordinates using image moments cv2.moments.
Missed
Passed
case-01 · gemini-3.6-flash · 8/23/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md403 linessingle file
OpenCV — Bio-image Computer Vision
Overview
OpenCV (cv2) provides optimized C++-backed image processing routines for preprocessing, segmentation, feature extraction, and video analysis of biological images. In life sciences, OpenCV is used for fluorescence image enhancement (background subtraction, CLAHE), morphological segmentation (watershed, contour detection), brightfield cell detection, and real-time microscopy stream processing. Unlike scikit-image (which emphasizes scientific measurement), OpenCV prioritizes computational speed and video support — making it ideal for preprocessing pipelines and real-time imaging applications.
When to Use
Preprocessing fluorescence or brightfield images: background subtraction, CLAHE, Gaussian/median blur
Detecting cell contours, blobs, or edges without deep learning (classical methods)
Processing video streams from live-cell imaging microscopes in real-time
Template matching for finding repeated structures (organelles, crystals, patterns)
Applying morphological operations (erosion, dilation, opening, closing) for mask refinement
Computing optical flow between video frames for cell tracking
Use scikit-image instead for scientific morphometry, regionprops, and scientific image I/O (TIFF metadata)
Use Cellpose or StarDist instead for deep-learning cell segmentation on fluorescence images
Prerequisites
Python packages: opencv-python, numpy, matplotlib
Optional: opencv-contrib-python for extra modules (SIFT, SURF, optical flow)