Claude Opus and Dome
Run Claude Opus 5.5 with failover, rules and a spend ceiling. Opus is the Claude you ration. It goes to the research agent working a hard problem, not the support bot answering a refund question. Dome pools it across Anthropic and AWS Bedrock, opens it to one group, and caps the bill.
Dome provides model brokeringand routing for Claude Opus.
One name, two providers
Agents send claude-opus-5-5, the pool's name. An Anthropic error moves that call to Bedrock, and the agent's request doesn't change.
Opus for the teams that need it
Tag a connection as frontier and a rule refuses it outside one group. The group comes from your identity provider.
A ceiling on spend
The pool gets $500 of provider spend a month and each caller 2M tokens a day, and both are enforced on the call.
Opus with failoverin five steps.
Two Opus connections, one pool, a group rule and a monthly cap.
Connect Anthropic
Tag the connection tier: frontier. The rule below reads that tag, and the Anthropic key stays in Dome's vault.
shelldome models add opus-55-anthropic \ --provider anthropic \ --model claude-opus-5-5 \ --api-key "$ANTHROPIC_API_KEY" \ --attributes '{"tier":"frontier"}' \ --gateway prod-gatewayConnect AWS Bedrock
Add a Bedrock connection in the dashboard with your AWS keys and region. Static AWS keys can only be set there.
Pool them for failover
Anthropic at priority 0, Bedrock at 1. Agents address the pool, so swapping a member later touches no agent code.
shelldome models pool create claude-opus-5-5 \ --failover-max all --gateway prod-gateway dome models pool member add claude-opus-5-5 opus-55-anthropic --priority 0dome models pool member add claude-opus-5-5 opus-55-bedrock --priority 1Point your agent at the Gateway
The research agent's client keeps its shape. Only the base URL and the key change, and the key is the agent's own.
pythonfrom openai import OpenAI client = OpenAI(base_url=f"{GATEWAY_URL}/v1", api_key=DOME_AGENT_KEY)client.chat.completions.create( model="claude-opus-5-5", messages=[{"role": "user", "content": "Summarize this incident."}],)Cap the spend
Quotas bind to the pool, a single connection, or each caller.
shelldome quotas set --subject pool --pool claude-opus-5-5 \ --unit provider_usd --limit 500 --window monthly dome quotas set --subject model --model opus-55-anthropic \ --per-caller --unit tokens --limit 2000000 --window daily
Opus for research,refused for everyone else.
Any connection tagged frontier is refused unless the person the agent acts for is in ml-research. The group arrives in their identity token.
cedar
forbid (principal, action == Dome::Action::"llm:invoke", resource is Dome::LLMModel)when { resource has tier && resource.tier == "frontier" }unless { principal has act_as && principal.act_as.groups.contains("ml-research")};One call,two outcomes.
Switch the caller or the argument and watch the same call decide differently, with every decision landing in audit.
Acting for
agent research-agent · acting as a.okafor
llm:invoke(model: "claude-opus-5-5")
- Agentresearch-agent is registered and active
- Callera.okafor verified, groups: ml-research
- RuleFrontier models are open to ml-research
- Quota$212 of $500 this month
DecisionAllowed
Bringing it alltogether.
Serving Claude Opus to registered agents, alongside their tools and identity, completes a governed agent application in Dome.
Acting for
Agent
Control point
GatewayEvery call decided and audited
RulesGuardsQuotas
Models
Claude OpusModel
AnthropicProviderCommon questionsabout Claude Opus.
More of whatDome works with.
Provider
OpenAI
The OpenAI API behind the Model Broker: the key held in Dome, every call authorized, metered and audited.
MCP server
GitHub
The GitHub MCP server behind the Tool Gateway, with rules that decide each call on its owner and repository.
Identity
Microsoft Entra ID
Entra access tokens verified on every agent call, so rules read app roles and audit names the person.
Runtime
OpenAI Agents SDK
OpenAI Agents SDK agents with their models 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.
AI models for enterprise agents: failover, rules and quotas
Every model your agents call goes through the Model Broker. Pool providers for failover, decide who may call each model, and cap what it costs.