AI Gateway

The Curate-Me AI Gateway is the control plane between your application and LLM providers. You keep your existing SDK, swap the base URL, add a Curate-Me key, and every request gets governance, routing, cost tracking, and auditability before it reaches the model.

The gateway sits upstream of agent harnesses — frameworks like LangChain, CrewAI, OpenAI Agents, and OpenClaw that orchestrate tool-calling agents. Regardless of which harness your team uses, the gateway applies consistent governance policies, tracks costs per harness, and provides a single audit trail across all of them.

What changes in your app

# Before
OPENAI_BASE_URL=https://api.openai.com/v1
 
# After
OPENAI_BASE_URL=https://api.curate-me.ai/v1/openai
X-CM-API-Key: cm_sk_xxx

You still send normal provider payloads. The gateway preserves standard JSON responses and streaming SSE behavior.

What the gateway adds

Endpoint patterns

Supported providers

50+ providers across built-in and custom routes:

See the full list on the Providers & Routing page.

Authentication

Every proxy request authenticates to Curate-Me with your gateway key (cm_sk_...). The upstream provider key is resolved separately:

Heads up: Don’t put a provider key (like sk-...) in the SDK’s api_key field when you rely on stored secrets — the SDK forwards it as Authorization: Bearer <provider-key>, which the gateway treats as an inline provider key and forwards upstream, shadowing your stored secret. Put your cm_sk_ gateway key in api_key instead, or send the provider key explicitly via X-Provider-Key.

Common response headers

Quick examples

OpenAI SDK

from openai import OpenAI
 
client = OpenAI(
    # Your Curate-Me gateway key — set it as api_key AND in the header.
    # Your OpenAI provider key is resolved from your stored Provider Secret.
    base_url="https://api.curate-me.ai/v1/openai",
    api_key="cm_sk_xxx",
    default_headers={"X-CM-API-Key": "cm_sk_xxx"},
)
 
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello"}],
)

Anthropic SDK

import anthropic
 
client = anthropic.Anthropic(
    # Your Curate-Me gateway key — set it as api_key AND in the header.
    # Your Anthropic provider key is resolved from your stored Provider Secret.
    base_url="https://api.curate-me.ai/v1/anthropic",
    api_key="cm_sk_xxx",
    default_headers={"X-CM-API-Key": "cm_sk_xxx"},
)
 
message = client.messages.create(
    model="claude-sonnet-4-5-20250929",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)

cURL

curl https://api.curate-me.ai/v1/openai/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-CM-API-Key: cm_sk_xxx" \
  -H "X-Provider-Key: $OPENAI_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Local development

cd services/backend
poetry install
poetry run uvicorn src.main_gateway:app --reload --port 8002

Then point your SDK at http://localhost:8002/v1/openai, http://localhost:8002/v1/anthropic, or another provider path.