NanoResearch CLI
An autonomous research assistant: from idea and plan through experiments and analysis to a paper, relying only on real results
- 1,333 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Generates and runs experiment code
- Calls external models and catalogs with a key
Install
- Works in
Manual install
pip install -e ".[dev]"Install from the cloned repository; then configure ~/.nanoresearch/config.json.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Generates and runs experiment code
- Calls external models and catalogs with a key
This is third-party code. Review the repository files before installing.
What it does
NanoResearch runs research in stages: ideation, planning, experiments, analysis and writing. Each stage is configured with its own model through an OpenAI-compatible API, with slash commands and a text interface. A key trait is that the writing stage uses only real results and artifacts, and where data is missing it writes a limitation rather than invented numbers. Literature can be searched through open catalogs, and the finished work can be exported.
Who it is for. For researchers and analysts who want a helper for the research cycle grounded honestly in results.
Good fit when
- You need to run research in stages from idea to text
- You need experiments and their analysis
- You need grounding in real results rather than invented numbers
Not a fit when
- You need a quick answer rather than a full research cycle
- You cannot run experiments that execute code
Example request
Run NanoResearch on this topic: gather ideas, plan experiments and prepare a paper draft from real resultsLimitations
It needs OpenAI-compatible model access set in a config file, whose availability from Russia should be checked separately. The experiment stages execute code, which needs resources and care.
How to disable. Remove the installed package via your Python package manager and delete the ~/.nanoresearch directory.
FAQ
Does it invent results?
No. The writing stage uses only real results and artifacts; where they are missing it writes a limitation.
Where does it search literature?
Through open catalogs such as OpenAlex; you can add your own key for higher rate limits.
Discussion
No questions or reviews yet.
Related
CLI toolsFeynman — a terminal AI agent for scientific research: literature reviews, source checks, paper critique and comparing a paper with its code
A research system on top of Codex CLI or Claude Code: runs of experiments, discussion of progress and a paper draft.
ARIS
ARIS (Auto-Research-In-Sleep)
A skill set for autonomous machine learning research: literature review, idea discovery, experiments and paper writing with review by a separate model
A set of 11 Claude Code skills that audits a third-party paper for self-consistency and fabrication and writes a report.
Awesome Autoresearch
awesome-autoresearch
A collection of autonomous self-improvement loops and research agents that set hypotheses and test them themselves
AI Research Skills
AI Research Skills Library
98 skills for ML research, from ideation and model training to inference, evaluation and paper writing
README in short
The README describes NanoResearch as an autonomous research assistant with ideation, planning, experiment, analysis and writing stages. Installation is from source via pip, model setup lives in a config.json for an OpenAI-compatible API, with environment variables and slash commands. It stresses that writing relies only on real artifacts, with no invented numbers. Literature is searched through open catalogs, and the finished work can be exported.