Will this skill change my documentation files?

No, the doc-classifier is strictly read-only and will never modify your files during the classification process.

What happens if a document type cannot be identified?

If both path and content analysis fail, the skill returns a result of '_untyped' with a baseline confidence score of 50.

How does doc-classifier determine the document type?

It uses a two-step process: first, it checks the file path for directory patterns; if that fails, it analyzes the file content for a 'fractary_doc_type' field in the frontmatter.

Which AI model powers this classification skill?

This skill utilizes the claude-haiku-4-5 model, which is optimized for fast and accurate text analysis tasks.

Documentation Classifier

Name: Documentation Classifier
Author: fractary

byfractary

•

Gestión de Contenido

Automatically identifies document types within a repository using intelligent path analysis and frontmatter content detection.

Acerca de

The Documentation Classifier skill streamlines the organization of project documentation by programmatically determining the type of any given file. It employs a multi-step strategy that first examines directory patterns for established conventions and then falls back to extracting metadata from file frontmatter. By providing a confidence score and the detection method used, it enables developers to build more structured documentation workflows and automated indexing systems within their development environment.

Características Principales

Deep content analysis via frontmatter metadata extraction
Optimized performance using the Claude Haiku 4.5 model
Granular confidence scoring (0-100) for every classification
2 GitHub stars
Prioritized path-based pattern matching for instant identification
Non-destructive, read-only operation to maintain file integrity

Casos de Uso

Validating repository structure against documentation standards
Pre-processing files for specialized AI-driven documentation search
Automatically categorizing markdown files for documentation portals

Acerca de

Características Principales

Deep content analysis via frontmatter metadata extraction
Optimized performance using the Claude Haiku 4.5 model
Granular confidence scoring (0-100) for every classification
2 GitHub stars
Prioritized path-based pattern matching for instant identification
Non-destructive, read-only operation to maintain file integrity

Casos de Uso

Validating repository structure against documentation standards
Pre-processing files for specialized AI-driven documentation search
Automatically categorizing markdown files for documentation portals