Github Research

This is a proper research tool that bridges academic papers and their implementations. Point it at a deep-research output directory and it runs a six-phase pipeline: extracts GitHub URLs and keywords from your research, searches multiple sources including Papers With Code, ranks 50-200 candidate repos, clones the top 15 for deep analysis, then produces an integration blueprint showing you which components are actually reusable. The script collection is extensive, all Python stdlib, covering everything from dependency extraction to cross-repo comparison matrices. It's built for the moment when you've read a dozen papers and need to know what code exists, what works, and how to combine it. Output is methodically organized with JSONL databases and markdown reports at each phase.

skills.sh installs: 1.8K
GitHub stars: 459
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add lingzhi227/agent-research-skills --skill github-research --agent claude-code

Installs into .claude/skills of the current project.

Files

GitHub Research Skill

Trigger

Activate this skill when the user wants to:

"Find repos for [topic]", "GitHub research on [topic]"
"Analyze open-source code for [topic]"
"Find implementations of [paper/technique]"
"Which repos implement [algorithm]?"
Uses /github-research <deep-research-output-dir> slash command

Overview

This skill systematically discovers, evaluates, and deeply analyzes GitHub repositories related to a research topic. It reads deep-research output (paper database, phase reports, code references) and produces an actionable integration blueprint for reusing open-source code.

Installation: ~/.claude/skills/github-research/ — scripts, references, and this skill definition. Output: ./github-research-output/{slug}/ relative to the current working directory. Input: A deep-research output directory (containing paper_db.jsonl, phase reports, code_repos.md, etc.)