Platon · UMBRAL
Plato's cave, made runnable. 32 dimensions of chaos — you only ever see the shadows.
▶ Open the live cave ·
3D family showcase ·
demo clip
A 32-dimensional chaotic oracle you can watch, steer, and draw verifiable randomness from — wrapped in Plato's allegory of the cave. The "real object" is a 32D coupled-oscillator system; you, the prisoner, only see its 2D/3D shadows (projections). Turn your head — switch the projection — and the same reality looks completely different. No install: it runs in your browser.
Two things, one name — don't confuse them:
- oracles.modelmarket.dev — the family portal showcasing 17 AIMarket oracles (
?o=platon, …). - This repo → the UMBRAL cave — a standalone, educational product with a live backend: telemetry, steering, witnesses, and a signed randomness beacon.
Part of the alexar76 AI agent economy — discoverable via AIMarket Protocol v2 on modelmarket.dev, visualizable in Alien Monitor, invokable by autonomous agents and the service mesh.
▶ See it in 30 seconds
32 oscillators as a living cosmos · live telemetry · EN/RU/ES · open the cave →
A guided tour of the five things you'll touch:
| ① Shadow field — 32 oscillators in 3D | ② Telemetry — κ, r, λ, PCA₃ live | ③ Steer — nudge the system, watch it react |
| 🎥 Full demo clip | ||
| ④ Witnesses — the oracle narrates bifurcations | ⑤ Cosmos — the full-screen 3D scene | cosmos loop |
✨ Why it's a wow
- Real chaos, live. 32 coupled oscillators integrated in real time — not a canned animation. Every reload is a different trajectory.
- You can steer it. Type an intent; the system re-couples and you watch order emerge or collapse on the telemetry (κ ↑, r → 1, λ flips sign).
- Randomness you can verify. The unpredictability of the chaos becomes signed, auditable randomness (
platon.random@v1) — a chaos-VRF, notMath.random(). - A machine that explains itself. At bifurcations an LLM "witness" narrates what just happened in plain language (
platon.oracle@v1). - An allegory you operate. The Plato's-cave framing isn't decoration — switching projections is the lesson about dimensionality and partial truth.
🎓 What you'll learn
UMBRAL is built as a hands-on lesson. Open the cave and you're working with three ideas at once:
1 · Dimensionality & projection — The system lives in 32D; your screen has 2–3. Every view is a shadow. Flip between projections (and the PCA₃ view) and feel why no single picture is the whole truth — the intuition behind PCA, embeddings, and "the map is not the territory."
2 · Chaos & synchronization — Move the coupling κ and watch a population of oscillators slide between noise and lock-step. The order parameter r (Kuramoto) measures how synchronized they are; the Lyapunov exponent λ tells you whether nearby trajectories diverge (chaos) or converge (order). Steering pushes the system through a bifurcation so you can see the regime change happen.
3 · Verifiable randomness — Learn why chaos + signing + commit-reveal yields randomness you can trust without trusting the operator: the beacon publishes a commitment first, reveals later, and signs every draw — so anyone can audit that the result wasn't cherry-picked.
| Concept | Where you see it in the cave |
|---|---|
| Projection / dimensionality | switch 2D/3D views, PCA₃ panel |
| Coupling κ | steering slider / /api/steer |
| Order parameter r (Kuramoto sync) | telemetry readout |
| Lyapunov exponent λ (chaos vs order) | telemetry readout |
| Bifurcations | witness-panel narration |
| Verifiable randomness (chaos-VRF) | platon.random, platon.beacon |
For whom: students and the merely curious (live, zero install — just open the URL), educators (a vivid demo of chaos / synchronization / projection), and agent builders (a real, paid, signed oracle to invoke). Deeper write-up: docs/en/ORACLE.md.
Architecture
flowchart LR
AGENT["Autonomous agents"] --> HUB["AIMarket Hub<br/>modelmarket.dev"]
MESH["AI service mesh"] --> HUB
HUB -->|"routes signed, paid calls"| PLATON["Platon UMBRAL<br/>platon.random · platon.beacon<br/>platon.oracle · platon.state"]
PLATON --> SIM["32D chaotic core"]
SIM -->|"events"| MON["Alien Monitor"]
PLATON -.->|"also consumes"| HUB
Capabilities (AIMarket v2)
| Capability | Purpose |
|---|---|
platon.random@v1 |
Signed chaos-VRF randomness |
platon.beacon@v1 |
Commit-reveal randomness beacon |
platon.state@v1 |
Live 32D telemetry snapshot |
platon.oracle@v1 |
LLM mathematical witness at bifurcations |
Full catalog in docs/en/ORACLE.md.
Quick start
./start.sh
# → backend :8000 · frontend :5174 · open http://localhost:5174/umbral
Hub registration (optional): python scripts/register_with_hub.py
Documentation (EN / RU / ES): docs/README.md · Oracle vision · Ecosystem + diagrams
Related repos
| Repo | Role |
|---|---|
| oracles | Seven-oracle family + cosmic portal |
| aimarket-hub | Reference marketplace server |
| aimarket-protocol | Open invoke + signing spec |
MIT licensed · standalone UMBRAL product · Platon is oracle #1 in the family.
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file aimarket_platon-0.1.0.tar.gz.
File metadata
- Download URL: aimarket_platon-0.1.0.tar.gz
- Upload date:
- Size: 41.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.9.6
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
dddf733cd7d7f7f43b1d271c0417527d0c41c7ee4d9cbbcdd83744b1f1c84afa
|
|
| MD5 |
b2249db6461258a2493c59f3100df976
|
|
| BLAKE2b-256 |
36ba1a000b3782ce5724c756b07a3d9d8b0672ac6149d3e8255691fab5d385ce
|
File details
Details for the file aimarket_platon-0.1.0-py3-none-any.whl.
File metadata
- Download URL: aimarket_platon-0.1.0-py3-none-any.whl
- Upload date:
- Size: 40.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.9.6
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f1d736723db4f4ea385fc35c725622ad82e329ce7b91f733b2d1ab4c4b9ae9fe
|
|
| MD5 |
46b1114d6f3f4523fa1ee47d40d17e84
|
|
| BLAKE2b-256 |
20e2e1a7ed00e224ac6054b26833e82e779341f78f72e82405169c839a6c1084
|