The sector's data, open to anything.

NTLSN is a static, public commons — no backend, no logins. But every dataset behind it is openly readable: query it, embed it, pipe it into a spreadsheet, or wire it straight into an AI assistant. One open source of truth for Australian higher-ed teaching & learning.

All endpoints return JSON over HTTPS with Access-Control-Allow-Origin: *, so they work from any site or script. Curation licensed CC BY-NC-SA 4.0; linked third-party works belong to their authors. NTLSN claims no endorsement by the universities or bodies it catalogues.

01The endpoints

Base URL https://www.ntlsn.com/data/. Everything is a plain static JSON file — fast, cacheable, and impossible to take down with traffic.

02Quick start — fetch it

No setup. From a browser, a Node script, a notebook — anywhere:

Copy// the next five sector events, anywhere
fetch('https://www.ntlsn.com/data/events.json')
  .then(r => r.json())
  .then(events => {
    const soon = events
      .filter(e => e.date >= new Date().toISOString().slice(0,10))
      .sort((a,b) => a.date.localeCompare(b.date));
    console.log(soon.slice(0,5));
  });

03Embed live events on your site

One line. A self-contained, shadow-DOM widget that never clashes with your CSS — sector-wide, or filtered to one university. ~5 KB, no dependencies.

Copy<script src="https://www.ntlsn.com/widget.js" data-limit="5" data-uni="usq"></script>

data-limit = how many to show (default 5). data-uni = optional university id (e.g. usq, unimelb, curtin) — omit for sector-wide. It reads /data/events.json and links back here.

04Wire it into AI — the MCP server

This is the part that turns NTLSN into infrastructure: a small Model Context Protocol server wraps these open endpoints into tools any AI assistant (Claude, etc.) can call — "what teaching conferences are coming up in WA?", "search the archive for work on academic integrity." MCP is the USB-C standard for AI ↔ data; NTLSN is the sector's plug.

Copy# ntlsn_mcp.py — expose NTLSN's open sector data to any AI assistant via MCP.
# pip install "mcp[cli]" httpx
import httpx
from mcp.server.fastmcp import FastMCP

BASE = "https://www.ntlsn.com/data"
mcp = FastMCP("ntlsn")

@mcp.tool()
def upcoming_events(uni: str = "", limit: int = 10) -> list:
    """Upcoming Australian higher-ed teaching & learning events. Optional uni id, e.g. usq."""
    data = httpx.get(f"{BASE}/events.json", timeout=15).json()
    today = __import__("datetime").date.today().isoformat()
    data = [e for e in data if e.get("date","") >= today]
    if uni: data = [e for e in data if e.get("uni") == uni.lower()]
    return sorted(data, key=lambda e: e["date"])[:limit]

@mcp.tool()
def search_archive(query: str, limit: int = 15) -> list:
    """Search the 1,431-work ALTC/OLT teaching-scholarship archive (1994-2025)."""
    q = query.lower()
    data = httpx.get(f"{BASE}/ltr.json", timeout=30).json()
    hits = [w for w in data if q in (w.get("t","") + " " + w.get("a","")).lower()]
    return hits[:limit]

@mcp.tool()
def universities() -> list:
    """All 42 Australian universities: L&T team pages, alliance group, Traditional Country."""
    return httpx.get(f"{BASE}/universities.json", timeout=15).json()

if __name__ == "__main__":
    mcp.run()

That's three tools over open data, no backend required. Fork it, add best_practices or recognition_framework, and you've given any AI the sector's teaching & learning memory. llms.txt describes the commons for AI crawlers too.

05Use it well

• Attribution: link back to ntlsn.com and keep the CC BY-NC-SA 4.0 notice. Non-commercial; share-alike.
• It's curated, not authoritative: the archive indexes titles/authors — full texts live at the Australian Government L&T Repository. Events are verified where the verified flag is true.
• No endorsement: NTLSN catalogues the sector; it doesn't claim partnership with the universities or bodies listed.
• Be kind: files are cached ~60 s. Don't hammer — fetch once and work locally.