What is GreenOps AI Studio?

GreenOps AI Studio is a multi-tier SaaS platform designed to proxy and optimize LLM workloads through an OpenAI-compatible AI API Gateway. It applies semantic caching on Cloudflare Workers, FinOps telemetry for token spend and carbon footprint, and a PII anonymization API before requests reach Gemini, OpenAI, or a local GPU runtime. Teams use it as a drop-in proxy alternative to calling provider APIs directly when they need cost control, privacy, and greener inference at the edge.

Introduction & architecture

GreenOps AI is an OpenAI-compatible LLM gateway on Cloudflare Workers. Every chat request is authenticated, PII-masked, optionally served from a semantic KV cache, then routed to Gemini, OpenAI, or a local GPU runtime.

Edge Worker

TypeScript proxy at the Cloudflare PoP. OpenAI-compatible POST /v1/chat/completions.

Semantic cache

WASM prompt hash + KV GREENOPS_CACHE. HIT replies skip the GPU and cost $0 / 0 g CO₂.

PII masking

Emails, cards, phones and tax IDs are anonymized before upstream. Count is returned on X-PII-Redacted-Count.

FinOps / ESG

Response headers expose USD saved, CO₂ avoided, cache HIT/MISS and user tier.

Cold path (MISS): trim + hash in WASM, call the LLM, write KV, emit FinOps headers.
Hot path (HIT): KV read only — typically tens of milliseconds on the same PoP.
Failover: Gemini 429/5xx automatically retries then falls back to OpenAI when configured.

Authentication & setup

Protected routes accept the gateway key as Authorization: Bearer <key> or the native header X-Proxy-Auth. For local and sandbox testing use the bootstrap key GreenOpsKey123.

Quick test key

Send Authorization: Bearer GreenOpsKey123 on chat, analytics and audit calls. Paid plans receive a gop_pro_… key after Stripe checkout. Provider keys (Gemini/OpenAI) are optional when the Worker already has server-side credentials.

Gateway authentication headers
Header	Required	Meaning
`Authorization`	Yes*	Bearer token. Example: `Bearer GreenOpsKey123`.
`X-Proxy-Auth`	Yes*	Equivalent gateway key if you do not send Authorization.
`X-User-Role`	No	`User` (default), `FinOps`, `Compliance`, `Admin`.
`X-User-Gemini-Key`	No	Bring-your-own Gemini key.
`X-User-OpenAI-Key`	No	Bring-your-own OpenAI key.
`X-Local-LLM-Url`	No	OpenAI-compatible local runtime (Ollama / GPU).
`X-Webhook-Url`	No	Slack / Teams incoming webhook for spend and security alerts.

Endpoints & API reference

Proxy URL: https://greenops-ai-proxy.copycomp-ai.workers.dev

Proxy URL

https://greenops-ai-proxy.copycomp-ai.workers.dev

POST · Chat Completions

https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/chat/completions

Proxy endpoints on the GreenOps AI gateway
Method	Path	Auth	Purpose
POST	`/v1/chat/completions`	Yes	OpenAI-compatible chat (JSON or SSE if stream: true)
GET	`/health`	No	Liveness + WASM engine
GET	`/v1/health`	No	KV / Gemini / OpenAI readiness
GET	`/v1/analytics`	Yes	FinOps / ESG totals
GET	`/v1/audit/logs`	Yes	EU AI Act audit trail (Compliance / Admin)
POST	`/v1/webhooks/stripe`	Stripe-Signature	Checkout → Pro key, cancel → Free
GET	`/v1/openapi.json`	No	OpenAPI 3.0 spec

Chat Completions

POST JSON with model + messages. Testing uses Bearer GreenOpsKey123. Swap in your provisioned key for production.

cURL

curl https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer GreenOpsKey123" \
  -d '{
    "model": "gemini-3.5-flash",
    "messages": [{"role": "user", "content": "Hello GreenOps"}]
  }'

JavaScript · Fetch

const res = await fetch('https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer GreenOpsKey123',
  },
  body: JSON.stringify({
    model: 'gemini-3.5-flash',
    messages: [{ role: 'user', content: 'Hello GreenOps' }],
  }),
});
if (!res.ok) throw new Error(`${res.status} ${await res.text()}`);
const data = await res.json();
console.log(data.choices?.[0]?.message?.content);

Python · Requests

import requests

url = "https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/chat/completions"
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer GreenOpsKey123",
}
payload = {
    "model": "gemini-3.5-flash",
    "messages": [{"role": "user", "content": "Hello GreenOps"}],
}
r = requests.post(url, json=payload, headers=headers, timeout=60)
if r.status_code == 401:
    raise SystemExit("Unauthorized: check Bearer GreenOpsKey123")
if r.status_code == 429:
    raise SystemExit(r.json())
r.raise_for_status()
print(r.json()["choices"][0]["message"]["content"])

Stripe webhooks

Point the Stripe Dashboard endpoint at the URL below. The Worker verifies Stripe-Signature when STRIPE_WEBHOOK_SECRET is set.

POST · Stripe webhook

https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/webhooks/stripe

checkout.session.completed issues an active Pro license (gop_pro_…, 100,000 req/month).
customer.subscription.deleted deactivates the license and returns the account to Free.
Configure the endpoint in Stripe as POST /v1/webhooks/stripe. No browser CORS is required.

cURL · webhook

curl -X POST https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/webhooks/stripe \
  -H "Content-Type: application/json" \
  -H "Stripe-Signature: t=1700000000,v1=…" \
  -d '{"type":"checkout.session.completed","data":{"object":{"customer_email":"dev@acme.com"}}}'

JavaScript · webhook

await fetch('https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/webhooks/stripe', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Stripe-Signature': 't=1700000000,v1=…',
  },
  body: JSON.stringify({
    type: 'checkout.session.completed',
    data: { object: { customer_email: 'dev@acme.com' } },
  }),
});

Python · webhook

import requests

requests.post(
    "https://greenops-ai-proxy.copycomp-ai.workers.dev/v1/webhooks/stripe",
    headers={
        "Content-Type": "application/json",
        "Stripe-Signature": "t=1700000000,v1=…",
    },
    json={
        "type": "checkout.session.completed",
        "data": {"object": {"customer_email": "dev@acme.com"}},
    },
    timeout=30,
)

Pricing tiers & quotas

Limits are enforced at the edge. Free is daily + burst; Pro and Business are monthly; Enterprise is unlimited.

GreenOps AI Studio pricing tiers
Plan	Price	Request limits	Features
Free	$0/mo	1,000 req/day · 30 req/min	Standard semantic caching, baseline PII masking, Community & Docs
Pro	$29/mo	100,000 req/month	Priority processing, real-time ESG/FinOps telemetry, automated PII anonymization
Business	$149/mo	1,000,000 req/month	ESG/FinOps PDF & CSV export, advanced PII rules, priority support
Enterprise	$499+/mo	Unlimited	Dedicated Workers & storage, custom ESG audit, 99.9% uptime SLA

Error handling & status codes

The gateway returns JSON errors. Chat quota responses include error.code, error.message and error.upgrade_url.

HTTP status codes returned by the AI gateway
Status	Name	Typical cause	What to do
401	Unauthorized	Missing/invalid Bearer or X-Proxy-Auth.	Send `Authorization: Bearer GreenOpsKey123` (test) or your provisioned key. Check for extra spaces.
429	Rate Limit Exceeded	Free 1,000/day or 30/min; Pro 100k/mo; Business 1M/mo.	Wait for `Retry-After`, or upgrade. Burst resets every 60 seconds.
500	Gateway Error	Unhandled Worker exception or upstream failure after retries.	Retry with backoff. Check GET /v1/health. Confirm Gemini/OpenAI keys if the model is cloud-hosted.

403: RBAC blocked the route (for example Audit requires Compliance/Admin) or prompt injection was rejected.
400: malformed JSON, missing messages[], or a required provider key is absent for that model.
SSE: send stream: true. Cache is bypassed on streaming responses.