Cli Qr Refiner

Professional utility to refine terminal-based ASCII QR code blocks (█, ▀, ▄) into high-definition PNG images.

Install

openclaw skills install @jdchi/cli-qr-refiner

CLI-QR-Refiner

A specialized tool for OpenClaw agents to capture and refine rough ASCII/Unicode QR code blocks from CLI terminal outputs into crystal-clear, scannable PNG images.

Features

Block Recognition: Automatically interprets █ (full), ▀ (upper), and ▄ (lower) block characters.
Pixel-Perfect Scaling: Renders vector-like precision to ensure zero scanning failures.
Zero Dependencies: Built on Python 3 + PIL (pre-installed on most systems).
Headless Optimized: Designed specifically for server-side environments where terminal displays are unreliable.

Dependencies

Runtime: Python 3 (no external packages required — uses built-in PIL/Pillow)

Tools & Commands

The refinement logic resides in {baseDir}/scripts/cli_qr_refiner.py.

Refinement Command

bash

python3 {baseDir}/scripts/cli_qr_refiner.py <input_txt_path> <output_png_path> [scale]

input_txt_path: Path to text file containing ASCII QR code
output_png_path: Path for output PNG image
scale (optional): Pixel size per character cell, default 10

Prompting / Behavior

When to use

Trigger this skill when:

A CLI tool (e.g., WeChat, Aliyun, NPM) outputs a block-character matrix intended for scanning.
The user reports scanning failures due to terminal alignment, line-height, or font issues.

How to use

Capture: Extract the raw character block from the terminal output.
Save: Write the captured block to a temporary file (e.g., /tmp/qr_source.txt).
Refine: Run the Python script to generate the high-def image.
Deliver: Send the refined PNG to the user.

Examples

Input (ASCII/Unicode)

text

██████████████  ██  ██████████████
██          ██  ██  ██          ██
██  ██████  ██      ██  ██████  ██

Output

A professional, high-definition PNG file, ready for any scanning app.

How it works

Each character in the ASCII QR maps to a cell:

█ → full black cell
▀ → upper-half black cell
▄ → lower-half black cell
(space) → white cell

The script scales each cell to scale × scale*2 pixels for crisp rendering.