claesbackman/AI-research-feedback

A collection of Claude Code skills for academic research review. These tools were developed by Claes Bäckman.

What it solves

This project provides a suite of specialized tools for academic researchers to get rigorous, AI-driven feedback on their work. It automates the tedious and high-stakes process of pre-submission reviews, ensuring that research papers, grant proposals, and pre-analysis plans are polished, internally consistent, and logically sound before they reach human referees.

How it works

The project consists of a collection of "skills" designed for Claude Code. These skills are installed as .md files in the Claude Code skills directory. When invoked via slash commands (e.g., /review-paper), they trigger specific multi-agent workflows. For example, the full paper review employs eight specialized agents—ranging from grammar checkers to "contribution skeptics"—who analyze LaTeX source files and consolidate their findings into a structured report.

Who it’s for

It is primarily designed for academic researchers, particularly those in economics and finance, who use LaTeX for writing and Stata, R, or Python for data analysis.

Highlights

  • Multi-Agent Review: Simulates a journal editorial board with specialized agents for style, consistency, mathematics, and adversarial critique.
  • Paper-Code Alignment: Checks if the empirical claims in a LaTeX paper are actually supported by the accompanying analysis code.
  • Grant & PAP Review: Dedicated workflows for reviewing grant proposals (e.g., NSF, NIH) and pre-analysis plans (PAPs).
  • Adversarial Auditing: Uses isolated subagents to perform "blind" audits of code changes to find errors without bias from the original author.
  • Educational Diffs: Converts code changes into HTML pages with plain-language explanations and quizzes to help coauthors understand the impact on results.

Related

  • Project
    luwill/research-skills

    A set of five Claude Code plug‑ins that automate academic research tasks—exhaustive literature search, citation‑integrity‑driven review writing, proposal drafting, and two kinds of slide‑deck generation (stylised images vs. fully editable, vector‑based decks). All tools enforce strict citation checks and are installable as simple skill folders.

  • Project
    pedrohcgs/claude-code-my-workflow

    A ready-to-fork foundation for AI-assisted academic work that uses Claude Code to automate the planning, creation, and verification of research papers, slides, and data analysis.

  • Project
    Yuan1z0825/nature-skills

    A library of 22 packaged AI-agent workflows ("skills") for academic research — paper reading, Nature-style writing, figure generation, reviewer responses, literature search, and patent drafting — installable into Claude Code, Codex, Chatbox, and other agents.

  • Project
    Boom5426/Nature-Paper-Skills

    A collection of 27 AI agent skills for Codex and Claude Code to help scientists write, revise, and manage rebuttals for Nature-series life science and computational biology papers.

  • Project
    michaelshimeles/skills

    A curated set of Claude Code agent skills that automate common development tasks—visual diff generation, service‑layer structuring, evidence‑driven UI testing, iterative Greptile review loops, work‑tree isolation, and prose clean‑up—plus a ready‑to‑drop workflow tying them together.