EurekaClaw Documentation#

The AI that catches your Eureka moments.
Crawls arXiv · Generates theorems · Proves lemmas · Writes LaTeX papers · Runs experiments

🚀 Quick Start

Up and running in 5 minutes. Install EurekaClaw, set your API key, and prove your first theorem.

Quick Start
📖 User Guide

Full walkthrough — input modes, gate mode, output files, tuning, troubleshooting, and example workflows.

User Guide
⚙️ Configuration

All .env variables: backends, models, token limits, pipeline modes, retry settings.

Configuration
🖥️ CLI Reference

Every command, option, and exit code for the eurekaclaw command-line tool.

CLI Reference
🐍 Python API

EurekaSession, KnowledgeBus, InputSpec, ResearchOutput, and all data models.

Python API
🏗️ Architecture

Pipeline stages, agent design, data flow, LaTeX compilation, and the theory inner loop.

Architecture

What EurekaClaw Does#

EurekaClaw is a multi-agent AI research assistant that goes from a question to a publishable result — autonomously. It crawls the literature, generates and stress-tests hypotheses, runs experiments, and writes up findings.

$ eurekaclaw prove "Find recent papers on sparse attention + prove efficiency bound"

🦞 Crawling arXiv cs.LG (2024–2025)...
📄 Found 23 relevant papers. Summarizing...
💡 Hypothesis generated: O(n log n) via topological filtration
✨ Theorem 3.1 drafted. LaTeX ready. Proof complete.
🦞 Eureka! Paper draft saved to ./results/
🔍 Literature Crawler

Fetch, summarize, and cross-reference papers from arXiv and Semantic Scholar.

💡 Idea Generator

Brainstorm novel hypotheses by synthesizing patterns across thousands of papers.

🔢 Theorem Prover

Generate, verify, and formalize proofs via a 7-stage bottom-up pipeline.

📄 Paper Writer

Draft camera-ready LaTeX papers with theorem environments and citations.

🖥️ Runs Locally

Use Ollama, vLLM, or any OpenAI-compatible endpoint — data stays private.

🧠 Continual Learning

Distills proof strategies into skills after every session, improving over time.

🧪 Experiment Runner (under development)

Numerically validates theoretical bounds; flags low-confidence lemmas.

🌐 Browser UI

Visual interface with live progress, settings sliders, and results viewer.

Installation#

git clone https://github.com/EurekaClaw/EurekaClaw
cd EurekaClaw
pip install -e "."
cp .env.example .env          # add ANTHROPIC_API_KEY

eurekaclaw install-skills     # install built-in proof skills (once)
eurekaclaw prove "The sample complexity of transformers is O(L·d·log(d)/ε²)" \
    --domain "ML theory" --output ./results

No API key? Use a Claude Pro/Max subscription via OAuth.

Documentation#

Getting Started

User Guide

Reference

Changelog

Acknowledgements#

EurekaClaw builds on ideas and inspiration from the broader AI-for-science community. We thank the authors of the following projects: