Can I use qmd for multiple different projects?

Yes, qmd supports 'collections' which allow you to index and search across different directories or projects independently using the --collection flag.

What is the difference between search and vsearch?

The 'search' command uses fast keyword-based matching (BM25), while 'vsearch' uses semantic similarity to find content that is conceptually related even if the exact keywords differ.

How does qmd save tokens?

Instead of Claude reading entire files to find information, qmd searches an index and provides only specific relevant snippets or file paths, reducing input volume significantly.

How do I keep the search index up to date?

You can use the 'qmd update' command to quickly re-index changed files or 'qmd embed' to refresh vector embeddings for semantic search.

Quick Markdown Search

Name: Quick Markdown Search
Author: mahmoudimus

bymahmoudimus

•

Content Management

Optimizes token usage by providing fast, hybrid search capabilities across local markdown documentation and developer notes.

QMD (Quick Markdown Search) is a specialized tool designed to help Claude navigate large codebases and documentation sets with maximum efficiency. By implementing a hybrid search system that combines BM25 keyword matching with semantic vector similarity, it allows users to pinpoint specific information without reading entire files upfront. This 'search-first' strategy significantly reduces token consumption—often by up to 90%—by identifying the most relevant file sections before they are loaded into the AI's context window, making it essential for exploring extensive technical knowledge bases.

Key Features

01Collection-based indexing for managing multiple documentation sets or project notes.

02Significant token optimization by retrieving only relevant snippets instead of entire files.

033 GitHub stars

04Hybrid search engine combining fast BM25 keyword matching and semantic vector similarity.

05Fast re-indexing and embedding updates to keep search results synchronized with local changes.

06Multiple search modes including path-only results and JSON output for easy integration.

Use Cases

01Navigating massive documentation libraries without exceeding AI context limits.

02Locating specific code patterns or functional logic across a distributed repository.

03Answering technical questions by querying a local markdown-based knowledge base.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mahmoudimus/simba qmd

For use in Claude.ai and ChatGPT

Download Skill