Azure OpenAI and Dome
Your Azure OpenAI deployments, governed per call. Connect each Azure OpenAI deployment to Dome with its endpoint, deployment name and API version. The key stays in Dome's vault, and agents call the Gateway with their own keys. Rules decide which people's work runs on which deployment.
Dome provides model brokeringand routing for Azure OpenAI.
Azure's api-key, kept by Dome
Dome sets the api-key header from its vault on every call. Agents authenticate with their own Dome keys.
EU work on EU deployments
Tag each deployment with its region. A rule keeps EU staff's calls on EU deployments.
Azure behind OpenAI, or the reverse
Pool a deployment with OpenAI's API for the same model, in either order. The second member covers the first.
Connect Azure OpenAIin three steps.
Add each deployment, pool it with OpenAI if you want cover, and let agents call the Gateway.
Add the deployment
The provider id is azure_openai. Endpoint, deployment and api_version are all required: Azure picks the model from the deployment.
shelldome models add gpt-azure-eu \ --provider azure_openai \ --model gpt-6-sol \ --provider-config '{"endpoint":"https://contoso-eu.openai.azure.com","deployment":"gpt-6-sol","api_version":"2024-10-21"}' \ --api-key "$AZURE_OPENAI_API_KEY" \ --attributes '{"region":"eu"}' \ --gateway prod-gatewayPool it with OpenAI
The EU deployment goes first and OpenAI second, and agents address the pool, gpt-6-sol.
shelldome models pool create gpt-6-sol \ --failover-max all --gateway prod-gateway dome models pool member add gpt-6-sol gpt-azure-eu --priority 0dome models pool member add gpt-6-sol gpt-openai --priority 1Point your agent at the Gateway
Use the standard OpenAI SDK, not the Azure client. The agent's Dome key replaces the Azure key.
pythonfrom openai import OpenAI client = OpenAI(base_url=f"{GATEWAY_URL}/v1", api_key=DOME_AGENT_KEY)client.chat.completions.create( model="gpt-6-sol", messages=[{"role": "user", "content": "Summarize this policy for a new hire."}],)
EU staffstay on EU deployments.
Calls made for anyone in eu-staff are allowed only on deployments tagged region: eu. Dome reads the group from that person's identity token.
cedar
forbid (principal, action == Dome::Action::"llm:invoke", resource is Dome::LLMModel)when { principal has act_as && principal.act_as.groups.contains("eu-staff")}unless { resource has region && resource.region == "eu" };Bringing it alltogether.
Connecting Azure OpenAI to registered agents, tools, and identity in Dome completes a governed agent application.
Acting for
Control point
GatewayEvery call decided and audited
RulesGuardsQuotas
Models
Common questionsabout Azure OpenAI.
More of whatDome works with.
Model
Claude Fable
Fable 5.1 from Anthropic and Amazon Bedrock in one failover pool, open to one group and capped by quota.
MCP server
GitHub
The GitHub MCP server behind the Tool Gateway, with rules that decide each call on its owner and repository.
Identity
Okta
Okta tokens verified on every agent call, so rules and audit name the person each agent acted for.
Runtime
LangGraph
LangGraph agents with their model calls on the Model Broker and their MCP tools on the Tool Gateway.
Client
Claude Code
Claude Code on a Dome Gateway with per-developer sign-in, rules on every tool call, and audit by name.
Agent service
TinyFish
TinyFish's web agents behind the Tool Gateway, with rules that decide each run on the site it targets.
Talk with ourFDE team.
Or start free, and register your first agent in minutes. Connect it through a gateway, give it rules, and see every call it makes in one audit trail.
Talk with our FDE team
Our engineers design, build and run governed agents for the part of your business you want to change. You see what they deliver.
LLM providers for AI agents: one governed path to every model
Connect a provider once. Its key stays in Dome, and every agent call to it is authorized, metered and audited.