Pywayne Cv Apriltag Detector

AprilTag corner detection for camera calibration and pose estimation. Use when working with pywayne.cv.apriltag_detector module to detect AprilTag fiducial m...

Security Scan

The instructions largely match an AprilTag detector, but there are gaps around runtime installation (use of an unspecified 'gettool') and an unstated dependency on the 'pywayne' package, which creates ambiguity about what will be downloaded and executed.

Purpose & Capability

The skill's name and description match the SKILL.md: it documents an ApriltagCornerDetector and expected inputs/outputs. However, the SKILL.md assumes the availability of the 'pywayne.cv.apriltag_detector' module but does not explain how 'pywayne' itself is installed or provided. That missing dependency is a usability/incoherence issue: either the environment must already include the pywayne package, or the skill should declare how to install it.

Instruction Scope

Instructions stay on-topic (detect AprilTags, draw results, support file paths and numpy arrays). They do instruct automatic installation of 'apriltag_detection' via 'gettool' at runtime, which broadens scope to include network downloads and package installation. The SKILL.md does not instruct reading unrelated files or exfiltrating data.

Install Mechanism

There is no formal install spec, but the SKILL.md says it will automatically install 'apriltag_detection' using 'gettool' if missing. 'gettool' is not described or pinned here: runtime automatic installers can fetch arbitrary code. Because downloads would happen at agent runtime and are not declared in an install spec, this elevates risk compared to a fully-declared, reviewable install step.

Credentials

The skill declares no required environment variables, credentials, or config paths. Nothing in the SKILL.md requests secrets or unrelated credentials.

Persistence & Privilege

The skill is not marked always:true and does not request persistent system-wide privileges. It is user-invocable and allows normal autonomous invocation; that default behavior is expected and not by itself concerning here.

What to consider before installing

This skill appears to implement AprilTag detection, but before installing or running it you should: (1) confirm where the 'pywayne' package comes from — the SKILL.md assumes that module exists but doesn't provide an install source; (2) treat the runtime auto-install via 'gettool' as potentially risky: determine what 'gettool' downloads (PyPI, GitHub releases, or an untrusted host) and whether packages are signed or pinned; (3) run the skill in an isolated environment (virtualenv, container) so any runtime downloads can't affect your system; (4) prefer skills that include an explicit, reviewable install spec or that name trustworthy package sources; and (5) if you need higher assurance, ask the publisher for the exact installation commands or a link to the source code for both 'pywayne' and 'apriltag_detection' before use.

Pywayne AprilTag Detector

This module detects AprilTag fiducial markers for camera calibration and pose estimation.

Quick Start

from pywayne.cv.apriltag_detector import ApriltagCornerDetector

# Create detector
detector = ApriltagCornerDetector()

# Detect from file path
detections = detector.detect('test.png', show_result=True)

# Detect from numpy array
import cv2
image = cv2.imread('test.png')
detections = detector.detect(image)

Detection Methods

detect()

Detect AprilTags in an image:

detections = detector.detect(
    image,           # File path, Path object, or numpy array
    show_result=False  # Show visualization window
)

Returns list of detection results with:

• id: Tag ID
• hamming_distance: Detection confidence
• center: Tag center coordinates (x, y)
• corners: 4 corner coordinates

detect_and_draw()

Detect AprilTags and draw results on original image:

result_image = detector.detect_and_draw(image)
cv2.imshow('Detection Result', result_image)
cv2.waitKey(0)

Visualization includes:

• Green polygon outlines
• Red corner circles
• Red ID labels at tag centers

Requirements

• cv2 (OpenCV) - Image processing
• numpy - Array operations
• gettool - Downloads apriltag_detection library automatically

Library Installation

The detector automatically checks for and installs the apriltag_detection library using gettool if not found.

Detection Result Format

Each detection contains:

Field	Description
id	Tag identifier
hamming_distance	Hamming distance (lower = more confident)
center	Tag center as (x, y) tuple
corners	4 corner coordinates as [(x1, y1), (x2, y2), (x3, y3), (x4, y4)]

Notes

• Supports both grayscale and BGR images
• Automatic grayscale conversion for detection
• Visualization sizes scale with image dimensions
• Uses AprilTag 36h11 tag family