migueltorrezd/os-taxonomy-fr — explained in plain English
Analysis updated 2026-05-18
Explore what French students learn at each age using the interactive 3D graph viewer.
Build educational tools or apps on top of structured, machine-readable curriculum JSON data.
Trace prerequisite relationships between school topics for lesson planning or research.
Use the included agent skill as a template to build a similar taxonomy for another country.
| migueltorrezd/os-taxonomy-fr | 100/talk_stock | alexcybernetic/playground-ai | |
|---|---|---|---|
| Stars | 2 | 2 | 2 |
| Language | HTML | HTML | HTML |
| Last pushed | — | 2022-03-24 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | easy | moderate | easy |
| Complexity | — | 3/5 | 1/5 |
| Audience | researcher | general | general |
Figures from each repo's GitHub metadata at analysis time.
Data is plain JSON, the viewer.html file works offline with no dependencies.
os-taxonomy-fr is a structured dataset that maps out what students in France are expected to learn, from early preschool through the final year of high school. It breaks the official French national curriculum into 6,051 small, single-idea topics, connects those topics with over 10,000 prerequisite links showing what a student needs to know before moving on, and ties everything back to nearly 14,500 curriculum standards taken word for word from official government documents. It is described as the French counterpart to an earlier US and UK curriculum project called os-taxonomy, and it uses a compatible data format on purpose. The project includes an interactive 3D viewer where every dot represents one topic, height on the graph represents the student's age, and connecting lines show prerequisite relationships. It has a bilingual English and French interface. A user can click on any topic to trace everything a learner needs to master beforehand, watch the curriculum unfold in age order, or switch the display into different visual layouts. A self-contained copy of this viewer ships in the repository and works offline with no setup. The data itself is plain JSON: one file for the micro-topics with English and French names, one for the prerequisite links, one for the raw curriculum standards grouped by subject family, one with parent-friendly summaries per subject and age range, and one listing the 127 official source documents the project was built from. The subjects span 42 areas across ten families, including mathematics, French language, science, history and geography, philosophy, and physical education. According to the README, the dataset was produced over three days using a pipeline of AI agents alongside code that enforced rules such as verifying quotes against the original text and preventing circular dependencies in the prerequisite links. The repository also includes a reusable recipe, built as a Claude Code agent skill, that walks through the same process so someone could attempt to build an equivalent taxonomy for another country. Licensing is split: the verbatim curriculum text is released under France's Open Licence v2.0, while the topics, prerequisite data, schemas, and viewer are released under the MIT license. The project is written mostly in HTML and would be most useful to researchers, educators, and developers who want structured, machine-readable curriculum data for France.
An open dataset that breaks France's national school curriculum into 6,051 linked micro-topics with an interactive 3D viewer.
Mainly HTML. The stack also includes HTML, JSON, Python.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.