paper-search

Search Markdown paper notes by title, author, keyword, field, or tag.

2|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/Clay-HHK/claude-config --skill paper-search-clay-hhk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: paper-search
Source: https://github.com/Clay-HHK/claude-config/tree/main/skills/paper-search
Command: npx skills add https://github.com/Clay-HHK/claude-config --skill paper-search-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps researchers and knowledge workers quickly locate relevant papers and matching passages within a curated collection of Markdown paper notes, removing the need to manually open and scan many files.

Core Features & Use Cases

  • Flexible search modes: supports title, author, keyword, field, and tag searches with case-insensitive matching.
  • Metadata extraction & ranking: extracts title, authors, date, and file path from frontmatter, captures matching context lines, and computes a heuristic relevance score based on title, content, author, field, and tag matches.
  • Organized output: groups results by research field and presents ranked entries including relevance, authors, date, link, and match location for quick review.
  • Use case: find all papers on "quantization" in the large-models folder, ranked by relevance and showing matching excerpts and links to notes.

Quick Start

Search the Papers directory for "large models quantization" and return ranked results with titles, authors, dates, file paths, relevance scores, and the matching context.

Frequently Asked Questions about paper-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search my Markdown paper notes to find relevant literature?

You can search Markdown paper notes by title, author, keyword, field, or tag. The tool performs a case-insensitive grep across frontmatter and content, then extracts matching text passages and metadata to locate relevant literature within your local knowledge base.

How does relevance ranking work for matching research papers?

Relevance ranking computes a heuristic score based on title, content, author, field, and tag matches. It evaluates case-insensitive grep results from Markdown frontmatter and content, then groups ranked entries by research field to organize the output for quick review.

Can I search for papers by specific authors and research fields in my notes?

Yes, you can query your Papers directory by author and field. The search extracts metadata from frontmatter and applies matching context extraction, allowing you to filter and locate specific research papers stored as Markdown notes in your local knowledge base.

What is the best way to find all papers on a specific topic in my local knowledge base?

The best way is to search the Papers directory using specific topic keywords. The tool applies exclusion filters and domain grouping, computing heuristic relevance scores to rank results and display matching excerpts with file paths and links to notes.

Does paper-search work with Markdown files that have frontmatter metadata?

Yes, it works with organized Markdown paper notes containing frontmatter. The tool performs case-insensitive grep across both frontmatter and content, extracting title, authors, date, and file path metadata to compute relevance scores and capture matching context lines.