CAJAL — Autonomous Agents

CAJAL is a local scientific paper generator that runs entirely on your machine, producing publication-ready papers with real citations from arXiv and CrossRef. It solves the problem of costly, cloud-dependent academic writing tools by offering a free, private, and decentralized alternative for researchers and developers. It is designed for anyone who needs to generate structured scientific papers, abstracts, or methodology sections without sending data to external servers.

🐙 View on GitHub ↗ Agnuxo1/CAJAL 📦 Install

✨ Key features

100% local computation, zero data exfiltration
Zero cost, MIT license, no subscriptions
Generates 7-section papers with real citations
Tribunal scoring by 8-10 LLM judges
100+ integrations for frameworks and IDEs
Works with any Ollama-compatible model

🎯 Use cases

Generate a full scientific paper from a topic
Create abstracts, methodology sections, or references
Review and score existing paper drafts
Integrate paper generation into agent frameworks
Automate paper creation in CI/CD pipelines

📦 Installation

🧰 Requirements: Python 3.8+, Ollama installed and running, and an Ollama-compatible model (e.g., llama3.1).

1. Install CAJAL

pip install cajal-p2pclaw

3. Create the CAJAL model

ollama create cajal -f integrations/ollama/Modelfile

4. Generate your first paper

python -c "from cajal_p2pclaw import PaperGenerator; \
  PaperGenerator().generate('Quantum error correction with surface codes')"

🚀 Usage

Generate a full paper

cajal generate "Federated learning for medical imaging privacy"

Generate only an abstract

cajal abstract "Neural architecture search for edge devices"

Generate methodology section

cajal methods "Differential privacy in distributed training"

Find references for a topic

cajal references "Byzantine fault tolerance in P2P networks" --count 12

Review an existing draft

cajal review draft.md

❓ FAQ

Does CAJAL require an internet connection?

No, CAJAL runs entirely on your machine. All computation is local, and no data leaves your computer.

What LLM models can I use with CAJAL?

CAJAL works with any Ollama-compatible model, such as llama3.1, qwen3.5, or mistral.

How does CAJAL ensure citations are real?

CAJAL integrates with arXiv and CrossRef to fetch verifiable references, avoiding hallucinated citations.

Can I use CAJAL from Python or JavaScript?

Yes, CAJAL provides a Python API and a JavaScript/TypeScript SDK via npm.

📊 Repository

GitHub github.com/Agnuxo1/CAJAL

Stars★ 21

Forks🍴 3

Open issues🐛 0

Last commit🕒 Jun 2, 2026

Created📅 May 2026

Language💻 Python

🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.