ArchestackBest of Open-Source AI: List on GitHub ↗3 stars (opens in a new tab)

#2424th of 28 in Agent platforms

autoresearchclaw

23-stage pipeline that turns a research topic into a LaTeX paper

Documentation ↗ (opens in a new tab)Repository on GitHub ↗ (opens in a new tab)
Stars
14.6k
License
MIT
Last commit
Aug 2026
Last release
May 2026
Language
Python

Missed the latest GitHub check; these facts may be older.

Overview

Runs a 23-stage pipeline from a topic to a paper. It searches literature via OpenAlex, Semantic Scholar and arXiv, runs sandboxed experiments, applies multi-agent review, and exports NeurIPS/ICML/ICLR LaTeX with BibTeX. It runs fully autonomous or with human approval in co-pilot modes, via CLI, Python API, OpenClaw, or ACP agents such as Claude Code.

Who it is for: Researchers wanting automated paper drafts from a topic, with optional oversight

Strengths

  • Outputs LaTeX, BibTeX, experiment code and charts in one deliverables folder
  • Citation verification checks arXiv, CrossRef, DataCite and an LLM
  • Six human-in-the-loop modes, from full-auto to step-by-step
  • Can use ACP agent CLIs as the LLM backend without API keys

Weaknesses

  • No Docker or compose files detected; setup is pip plus an interactive setup command
  • Needs an LLM provider or agent CLI; quick start uses an OpenAI key
  • Domain executors such as HEP via Magnus cloud add external dependencies
  • RAM and GPU requirements are not stated in the README

What it needs

  • GPU optional
  • Needs Python 3.11+, LLM API or ACP agent CLI, OpenAlex / Semantic Scholar / arXiv, LaTeX, Docker (optional)
  • Models: OpenAI-compatible APIs, ACP agents (Claude Code, Codex CLI, Gemini CLI, Copilot CLI, Kimi CLI, OpenCode)

Also in Agent platforms

See all 28
Also in Agent platforms
RankProjectScore
1LangflowVisual builder for AI agents and workflows, exposed as API or MCP server155.7k starsMIT81 out of 100
2daily_stock_analysisDaily LLM stock analysis for A-share, HK, US and other markets, pushed to chat apps66.4k starsMIT80 out of 100
3mindshubAgent workspace for knowledge work, with swappable open agent harnesses Live demo ↗ (opens in a new tab)39.8k starsMIT78 out of 100
#4deer-flowAgent harness that runs sub-agents, memory and sandboxed skills for long tasks83.9k starsMIT76 out of 100
#5DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs Live demo ↗ (opens in a new tab)158.5k starscustom license75 out of 100