# PrathamLearnsToCode/paper2code

Agent skill to turn any arxiv paper into a working implementation

Repository: https://github.com/PrathamLearnsToCode/paper2code
Canonical: https://ross.abutalabs.com/products/prathamlearnstocode-paper2code
Language: Python
License: MIT
License Family: permissive
Topics: agent, claude-code, skills
Last push: 2026-04-03T13:03:52+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 75, release rhythm 35, longevity 10
- inputs: {"age_days": 152, "days_push": 152, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1509, forks 176 (observed 2026-08-28T04:04:55.635741+00:00)

## What it is
An agent skill for coding agents like Claude Code that converts an arXiv paper URL into a working Python implementation. It generates citation-anchored code where every line references the paper section it implements, and flags unspecified implementation choices instead of silently inventing them.

## Use cases
- turn an arxiv paper into working code
- reproduce a machine learning paper implementation
- generate code from a research paper with citations
- audit which parts of a paper implementation are unspecified
- implement a paper's model and training loop automatically
- find hyperparameters buried in paper appendices

## When to choose
- you need to reproduce an ML paper and want traceable, citation-anchored code
- you use Claude Code or compatible coding agents and want paper-to-code automation
- you want ambiguity in a paper's method explicitly flagged rather than guessed

## When to avoid
- you need a guaranteed runnable, verified implementation without manual review
- your target paper has no arXiv version or is not machine-learning focused
- you want a standalone application rather than an agent skill

## Facets
- artifact type: plugin
- maturity: active
- function: agent-framework, llm-inference, code-review, developer-tools
- domain: machine-learning, large-language-models, developer-tools, artificial-intelligence
- platform: python, cli, cross-platform
- tags: claude-code, agent-skill, paper-reproduction, arxiv, citation-anchoring, code-generation

## Member repositories
- PrathamLearnsToCode/paper2code (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:55.635741+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:32:28.526635+00:00, confidence not recorded.
  - readme: https://github.com/PrathamLearnsToCode/paper2code (fetched 2026-08-28T04:04:55.635741+00:00, sha 0a8a3c4c790c)
- Data as of 2026-08-30T08:39:29.467469+00:00.