Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
paper2code reads an arxiv paper, identifies its core contribution, and produces a minimal Python implementation anchored to the original source. It runs an ambiguity audit to flag every detail the paper leaves unspecified, then writes code only for what is stated. The result is a reproducible baseline that researchers, students, and engineers can inspect, extend, or compare against.
WHO IT'S FOR
ML Research Engineers
rapidly prototype paper implementations for experiments
ML PhD Students
reproduce core contributions for research validation
Applied AI Teams
evaluate new models by generating minimal reference code
ML Educators
create paper-based assignments with verifiable code
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
Install with npx
Installs: Global (recommended) or project-level, selected interactively · Claude Code
In a terminal, run the installer. When prompted, select Claude Code as the coding agent and choose either Global (recommended) or project-level scope; then choose Symlink (recommended) or copy.
Claude