Dashboard
Monitor your AI models · GOLD epistemic layer · cryptographic proof chain
Plan
OPEN
10/day
Status
Active
—
Requests today
0
today
/ —
ONTO API Key
Score History
LOCAL
Epistemic Metrics
— last test · GOLD discipline layer
● GOLD Active
QD
—
Quantification
density score
SS
—
Source Signals
cited refs
UM
—
Uncertainty
acknowledged
CP
—
Counterargs
opposing views
EpCE
—
GOLD deviation
vs baseline
Quick Test
All tiers
Send a query through GOLD
● GOLD Active
Select a connected model → ask a question → GOLD discipline layer is injected automatically → response is scored and cryptographically signed.
⚠ AI Provider: proxy is demo only. Production delivery via SSE with zero latency.
Score
—
Grade
—
Proof
—Score Any Text
All tiers
Paste any AI response → measure epistemic discipline
Paste an AI-generated response below. GOLD scoring engine will measure quantification density, source specificity, uncertainty marking, and more.
Score
—
Grade
—
Risk
—
Compliance
—
Proxy— your app routes through ONTO, GOLD injected per request
Testing
No proxy models yet
0 / 10 req/day
Sampled verification
SSE Requires Provider Tier
Upgrade to enable continuous verification
SSE— GOLD cached on your server, zero latency, continuous push
Production
No SSE models yet
∞ Unlimited
Continuous verification
Integration
3 lines to connect your app — add x-api-key header, GOLD does the rest
View integration code & examples
Connect via SSE, receive GOLD, inject into your models. Zero latency overhead.
How it works
1. Connect to SSE → receive GOLD discipline layer
2. Cache GOLD locally → inject into system prompt
3. Your models serve requests directly — ONTO is not in the request path
4. SSE heartbeat every 30s. GOLD updates pushed automatically.
2. Cache GOLD locally → inject into system prompt
3. Your models serve requests directly — ONTO is not in the request path
4. SSE heartbeat every 30s. GOLD updates pushed automatically.
Step 1 — Connect to SSE
# Connect to ONTO SSE — receive GOLD discipline layer
import requests, json, threading
ONTO_KEY = "YOUR_ONTO_API_KEY"
GOLD = "" # Updated by SSE listener
def sse_listener():
global GOLD
r = requests.get(
"https://api.ontostandard.org/v1/gold/stream",
headers={"x-api-key": ONTO_KEY},
stream=True
)
for line in r.iter_lines():
if not line: continue
data = json.loads(line.decode().removeprefix("data: "))
if data["type"] in ("gold_corpus", "gold_update"):
GOLD = data["discipline_layer"]
print(f"GOLD received: {len(GOLD)} chars")
threading.Thread(target=sse_listener, daemon=True).start()Step 2 — Inject into your models
# Inject GOLD into every request — your existing code
from openai import OpenAI
client = OpenAI() # Your own API key, your own infra
def chat(user_message):
return client.chat.completions.create(
model="gpt-4o",
messages=[
{"role": "system", "content": GOLD}, # ← from SSE
{"role": "user", "content": user_message}
]
)
Zero overhead. ONTO is not in your inference path. Your models serve millions of requests directly. GOLD updates arrive via SSE — no polling, no proxy, no latency.
Python, cURL — 3 lines of code. Your ONTO API key is pre-filled.
How it works
Your app → ONTO Proxy → AI Provider (OpenAI, Anthropic, etc.)
ONTO injects GOLD discipline layer into every request. Your provider key never stored.
ONTO injects GOLD discipline layer into every request. Your provider key never stored.
Python (OpenAI-compatible)
# Replace your OpenAI client with:
client = OpenAI(
api_key="YOUR_ONTO_API_KEY",
base_url="https://api.ontostandard.org/v1/proxy",
default_headers={
"X-Provider-Key": "sk-...your_openai_key"
}
)cURL
curl -X POST https://api.ontostandard.org/v1/proxy/chat/completions \
-H "x-api-key: YOUR_ONTO_API_KEY" \
-H "X-Provider-Key: sk-...your_openai_key" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-4o","messages":[{"role":"user","content":"Your question"}]}'NEXT
—
✓ PROXY
✓ GOLD
✓ SIG
✓ CHAIN
ONTO does not guarantee accuracy. Not responsible for decisions made by AI systems or their operators.
API & SDK
Connect your AI to ONTO GOLD epistemic discipline
API Key
Quick Start
Python — OpenAI-compatible (3 lines changed)
# Before (direct to OpenAI):
client = OpenAI(api_key="sk-...")
# After (through ONTO proxy — GOLD injected server-side):
client = OpenAI(
api_key="onto_sk_...",
base_url="https://api.ontostandard.org/v1/proxy",
default_headers={"X-Provider-Key": "sk-...your_openai_key"}
)
# Everything else stays the same
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": "Your question"}]
)Other providers (DeepSeek, Groq, Mistral, etc.)
# Same endpoint, add X-Provider-Base-URL header:
client = OpenAI(
api_key="onto_sk_...",
base_url="https://api.ontostandard.org/v1/proxy",
default_headers={
"X-Provider-Key": "sk-...your_provider_key",
"X-Provider-Base-URL": "https://api.deepseek.com/v1"
}
)cURL
curl -X POST https://api.ontostandard.org/v1/proxy/chat/completions \
-H "x-api-key: onto_sk_..." \
-H "X-Provider-Key: sk-...your_openai_key" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-4o","messages":[{"role":"user","content":"Your question"}]}'
How it works: Your request → ONTO injects GOLD discipline into system prompt → forwards to your provider → returns enhanced response. ONTO never stores your prompts or responses. Standard provider token costs apply.
Rate Limits
| Tier | Proxy Rate | GOLD Level | Proof Chain |
|---|---|---|---|
| Open | 10/day | GOLD Core | Ed25519 signed |
| Standard | 1,000/day | GOLD Extended | Ed25519 signed |
| AI Provider | Unlimited | GOLD Full | Ed25519 + audit trail |
Current Plan
Open · 10 requests / day
0 / 50 proxy calls used today
Account
Identity
Name
John Doe
Edit
Email
support@ontostandard.org
Type
Individual
Organization
No organization registered. Required to generate invoices and access institutional features.
Name
—
Edit
Entity Type
—
Industry
—
Registration
—
VAT
—
Address
—
Status
Verified
Subscription
No active subscription. Using Open tier (10 evaluations/day).
Active Plan
—
Rate Limit
—
Plan Expires
—
Total Paid
—
Documents
Specifications & Policies
View
Support
General
support@ontostandard.org
Sales
support@ontostandard.org
Billing
support@ontostandard.org
Access Tiers
Open
Free
Evaluate ONTO on your AI before you commit
- 10 proxy requests/day
- GOLD epistemic discipline layer
- Scoring API
Standard
$2,500/month
or $30,000/year
Production GOLD discipline for companies using AI
- 1,000 proxy requests/day
- GOLD epistemic discipline layer
- Scoring API
AI Provider
$250,000/year
Provider-native SSE evaluation interface
- Direct event-stream integration
- No proxy replication layer
- No traffic amplification
- GOLD epistemic discipline layer
- Scoring API
SSE integration · fixed annual license · unlimited scale · ONTO not in inference path
Grant Program
Free Standard access for independent researchers, early-stage startups, and nonprofit organizations whose work benefits society. Grant recipients receive the same infrastructure as paid Standard customers: 1,000 proxy requests/day, GOLD Extended, signed proof chain. No restrictions, no watermarks.
Invoice payment requires registered organization. Account settings
Invoice
No active subscription
Reference
Admin panel