How do I handle repositories that fail auto-detection?

If a project type isn't recognized, you can use the --mask argument to manually provide a glob pattern for the file types you wish to index.

What happens if a repository already exists in my refs folder?

The skill is idempotent; if the repository exists, it will attempt a fast-forward pull to sync with the remote instead of re-cloning.

What is the purpose of the QMD Add skill?

It automates the process of downloading GitHub repositories to a specific local directory (~/Developer/refs/), indexing their contents based on file type, and creating embeddings for semantic search.

Which programming languages does it automatically detect?

The skill currently supports auto-detection for TypeScript, Rust, Go, Python, and Swift by looking for specific manifest files like package.json or Cargo.toml.

Can I use this skill if I don't want to download the whole Git history?

Yes, by default the skill performs a shallow clone with a depth of 1. You can use the --full flag if you require the complete commit history.

QMD Repository Indexer

Name: QMD Repository Indexer
Author: ramonclaudio

byramonclaudio

学习与文档

Clones, indexes, and embeds GitHub repositories into a local knowledge base for semantic search and AI-assisted analysis.

关于

The QMD Add skill automates the entire workflow of bringing external codebases into your local development environment for advanced search and context. It handles repository cloning (shallow or full), intelligently detects project languages to apply optimized indexing filters, and generates vector embeddings to enable semantic search capabilities. By managing collections within the QMD ecosystem, it ensures that your external references stay synchronized with their remote sources while providing the AI with high-quality, structured context from READMEs and source files.

主要功能

Seamless collection management with custom indexing masks
Automated GitHub repository cloning and fast-forward updates
0 GitHub stars
Intelligent project type detection for TS, Rust, Go, Python, and Swift
Vector embedding generation for high-performance semantic search
Idempotent execution with dry-run support for safe planning

使用场景

Rapidly indexing open-source projects to answer complex implementation questions
Building a local searchable library of third-party documentation and reference code
Maintaining a synchronized set of 'ref' repositories for architectural patterns

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ramonclaudio/skills add

For use in Claude.ai and ChatGPT

Download Skill

GitHub

关于