Gemini on Vertex AI and Dome
Gemini on Vertex AI, with the service account kept in Dome. Dome reaches Gemini through Vertex AI in your Google Cloud project. The service account key lives in Dome's vault, and Dome trades it for short-lived Google tokens on each call. Agents authenticate as themselves, and rules decide who gets which model.
Dome provides model brokeringand routing for Google Gemini.
The service account stays in Dome
The JSON key sits in Dome's vault. Dome exchanges it for short-lived Google tokens, and agents never see either.
Models by team
Tag each Vertex connection with its tier. A rule keeps the support team on Flash.
One project, many agents
Every call is metered against the agent and the person it acted for. Spend is attributed per agent, not per project.
Connect Vertex AIin two steps.
Add a connection per Gemini model with your project and location, then point agents at the Gateway.
Add the connection
The provider id is google. Pass the service account's JSON key as the credential; location defaults to us-central1.
shelldome models add gemini-flash \ --provider google \ --model gemini-3.8-flash \ --provider-config '{"project":"acme-agents","location":"us-central1"}' \ --api-key "$(cat vertex-sa.json)" \ --attributes '{"tier":"flash"}' \ --gateway prod-gatewayPoint your agent at the Gateway
Any OpenAI-compatible client works. Dome translates each request to Gemini's format and back.
pythonfrom openai import OpenAI client = OpenAI(base_url=f"{GATEWAY_URL}/v1", api_key=DOME_AGENT_KEY)client.chat.completions.create( model="gemini-flash", messages=[{"role": "user", "content": "Summarize this support thread."}],)
Supportstays on Flash.
Refuses connections tagged pro when the person the agent acts for is in support, while other teams can use either.
cedar
forbid (principal, action == Dome::Action::"llm:invoke", resource is Dome::LLMModel)when { resource has tier && resource.tier == "pro" && principal has act_as && principal.act_as.groups.contains("support")};Bringing it alltogether.
Connecting Google Gemini to registered agents, tools, and identity in Dome completes a governed agent application.
Acting for
Agent
Control point
GatewayEvery call decided and audited
RulesGuardsQuotas
Models
Google GeminiProvider
GeminiModelCommon questionsabout Google Gemini.
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.