Logo Lanfrica

atwine/fair-maturity-copilot

Domaine:

digital infrastructure

Type de record:

tools
Créateur:
atw
Hôte:
Guided, plain-language FAIR data-maturity self-assessment — built as a reusable standard-to-plain-language engine (ACE Uganda) fair-maturity-copilot A guided, plain-language tool that helps research organizations assess how "FAIR" their data is — Findable, Accessible, Interoperable, and Reusable — without needing a data librarian on staff. You answer plain-language questions about your dataset, get a score, and the tool turns every weak spot into a specific, actionable next step. ## What it does **For a single dataset** — 12 plain-language questions based on the RDA FAIR Data Maturity Model. You answer each one (with worked examples shown alongside), get a score out of 100, and a plain-language report explaining what's working and what to fix — with concrete steps for each gap. It also generates an ordered action plan and offers a chat-based mentor to coach you through fixing things one step at a time. **For multiple sites in one initiative** (e.g. a multi-center consortium) — 6 questions about whether your sites describe their data consistently enough to combine and compare: shared field names, a common data dictionary, standardized categories, a way to link records across sites. A "we haven't started this yet" answer is a real option here — it doesn't count against your score, because being early is normal, not a failure. See the in-app `/about` page (or `docs/WHY-THIS-TOOL.md`) for how this tool fits alongside other FAIR resources like F-UJI, the FAIR Cookbook, and FAIR-DSM. ## Quick start You'll need **Python 3.12+**, **Node.js 18+**, and an **LLM provider** (see below). The app runs as two servers: a Python backend and a Next.js frontend. ### 1. Set up the backend ```bash cd backend python -m venv .venv ./.venv/Scripts/python.exe -m pip install -e ".[dev]" # Windows; on macOS/Linux use .venv/bin/python cp .env.example .env ``` Now edit `backend/.env` and fill in two things: - **`DATABASE_URL`** — a Postgres connection string. The easiest option is a free Neon database (takes 2 minutes to set up). For quick local testing you can also use `sqlite:///./dev.db`. - …