What types of garbled text can this skill fix?

It supports 8 common types including 'Kun-Kao' (UTF-8/GBK mismatch), symbol-based corruption (ISO8859-1), pinyin codes, and box/question mark characters.

How accurate is the repair process?

The skill uses a scoring system that validates Chinese character frequency, only returning results with a confidence score higher than 0.6 to ensure readability.

Is file batch processing supported?

Yes, the skill includes scripts that allow you to specify input and output file paths for automated batch repair.

Can it fix 'Tang Tang Tang' (烫烫烫) errors?

No, 'Tang' and 'Tun' are memory initialization errors in C/C++, but the skill can identify them and advise on checking source code.

Do I need to know the original encoding type?

Not necessarily. The tool intelligently detects the most likely encoding mismatch, though providing a hint can improve the success rate for complex cases.

Chinese Garbled Text Fixer

Name: Chinese Garbled Text Fixer
Author: yantyx

byyantyx

Developer Tools

Detects and repairs common Chinese text encoding issues (mojibake) like GBK/UTF-8 mismatches and memory-related artifacts.

About

This skill provides specialized capabilities for Claude to intelligently identify and restore garbled Chinese characters (Mojibake) caused by encoding mismatches or uninitialized memory. It supports eight distinct types of corruption—including the notorious 'Kun-Kao' (锟拷) and 'Tang-Tang-Tang' (烫烫烫)—by offering automated detection and repair scripts. Whether dealing with legacy databases, cross-platform log files, or misconfigured text editors, this tool applies the correct conversion logic (GBK, UTF-8, ISO8859-1) to recover readable content with high confidence.

Key Features

Intelligent detection of 8 major Chinese garbled text patterns
Confidence-based validation system (minimum 0.6 score) to ensure text accuracy
Support for manual encoding hints (e.g., ISO8859-1, GBK) to accelerate recovery
Diagnostic identification for unrecoverable memory errors like 'Tang' and 'Tun'
Automated repair for both individual strings and batch file processing
0 GitHub stars

Use Cases

Identifying encoding-related bugs in source code versus memory-related initialization issues
Fixing corrupted log outputs or database strings containing 'Kun-Kao' (锟拷锟斤拷) artifacts
Recovering readable content from legacy text files showing unreadable 'Guwen' characters

About

Key Features

Intelligent detection of 8 major Chinese garbled text patterns
Confidence-based validation system (minimum 0.6 score) to ensure text accuracy
Support for manual encoding hints (e.g., ISO8859-1, GBK) to accelerate recovery
Diagnostic identification for unrecoverable memory errors like 'Tang' and 'Tun'
Automated repair for both individual strings and batch file processing
0 GitHub stars

Use Cases

Identifying encoding-related bugs in source code versus memory-related initialization issues
Fixing corrupted log outputs or database strings containing 'Kun-Kao' (锟拷锟斤拷) artifacts
Recovering readable content from legacy text files showing unreadable 'Guwen' characters