Getting Started

Getting Started

Everything runs from a Makefile; the quality gate (make check) must be green before any commit.

Prerequisites

• Python 3.10+ (a venv is created for you), and Node.js (for the map build).
• No API key needed to build; a key (OpenAI or any model CLI) is only needed for the live eval.

Build it

make setup      # venv + install runtime & dev deps
make fibo       # fetch the pinned FIBO source (sparse checkout of all 10 domains)
make commons    # fetch the OMG Commons upper ontology (CMNS)
make all        # extract FIBO -> intermediate.json -> knowledge/ OKF bundle -> validate
make curate     # nominate core + bridges for all use cases, rebuild the bundle
make map        # knowledge/ + okf.config.js -> js/data.js (the Cytoscape map data)

Install it as an app (PWA)

Kuber Map is an installable Progressive Web App: it works offline and opens in its own window.

• On your phone — scan this code (or open the URL), then use Add to Home Screen (iOS Safari: Share -> Add to Home Screen; Android Chrome: menu -> Install app):

  

  https://ai-first-community.github.io/kuber-map/

• On desktop (Chrome/Edge) — open the app and click the install icon in the address bar.

Open the map

Serve the repo root over http (so the offline/PWA service worker works) and open index.html (the landing) → Enter the map (app.html). The map lands on the curated core (284 concepts across five use cases). Use the Use-case lens to focus one, or toggle Core → full ontology for all 3,104. Click any node for its definition, examples, connections, and FIBO IRI citation.

Context packs and the contribution

make pack       # export every use case's context pack -> export/<use-case>/
make contrib    # package the 19 cross-domain bridges as an EDM proposal -> contrib/

Run the eval

make eval       # offline mechanism check (deterministic, no model/key)

# live value proof against any use case (needs a model command):
export OPENAI_API_KEY=sk-...    # or put it in .env (gitignored); openai_cli auto-loads it
EVAL_LLM_CMD='python eval/openai_cli.py' python eval/harness.py --adapter llm --model gpt-4o-mini \
  --benchmark eval/kyc-benchmark.json --pack export/kyc/pack.json

See Value Proof for the current numbers, and the in-repo docs/Getting-Started.md for the fuller version.