Skip to content

About

Local scientific AI workbench with tools, notebooks, model adapters and provenance.

Resources

Security policy

Stars

2 stars

Watchers

0 watching

Forks

Repository files navigation

Superscience workflow overview

Superscience

A local research workbench with tools, notebooks and evidence.

License: Apache-2.0 Validation notes

Overview · Quick start · Workflow · Validation · Docs

English · 简体中文


Overview

A Go/Wails/React research workspace for projects, assistant sessions, files, PTY terminals, notebooks, scientific search and inspectable artifacts. Inspired by the OpenScience product boundary; this repository contains its own implementation.

Capability What it provides
Local tools File editing, real terminals, permissions and persistent sessions.
Science workspace Provider adapters, scientific connector catalog, notebook and LSP integrations.
Evidence Artifacts, provenance and optional self-hosted sharing.

Local demo

superscience local demonstration

First-run provider setup; no external model request was made.

Quick start

git clone https://github.com/eust-w/superscience.git
cd superscience
# Requires Go 1.25+, Node.js 24 and pnpm 11.7.0.
pnpm install --frozen-lockfile
pnpm --dir frontend build
go run ./cmd/superscience serve
# Open http://127.0.0.1:4096/

First launch requires a provider/model configuration and authenticated verification before entering the workspace. E2E uses a localhost protocol fixture; it does not establish acceptance of a real cloud model.

Workflow

INVESTIGATE → EXECUTE → TRACE
python3 -m venv .venv
.venv/bin/python -m pip install -r workers/python/requirements.txt
.venv/bin/python -m ipykernel install --prefix "$PWD/.venv" --name python3
export SUPERSCIENCE_PYTHON="$PWD/.venv/bin/python"
export JUPYTER_PATH="$PWD/.venv/share/jupyter"
make verify

Validation

make verify

Recorded environment and executed checks: validation notes.

Area Current boundary
Providers Adapters and protocol fixtures do not establish live acceptance of every provider.
Notebooks / LSP Require Jupyter kernels and installed language servers. Permission controls are not a process sandbox.
Distribution Desktop signing, cloud deployment and external model credentials are separate setup steps.

Docs

License and provenance

Original code and newly created demo assets are licensed under Apache-2.0. Dependencies, models, datasets and external resources retain their terms; see THIRD_PARTY_NOTICES.md.

About

Local scientific AI workbench with tools, notebooks, model adapters and provenance.

Resources

Security policy

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages