In your VPC, your way
Deploy the gateway with Helm or Terraform in AWS, Azure or Google Cloud, in your VPC or Kubernetes cluster.
- Helm or Terraform
- AWS, Azure or Google Cloud
- No agent rewrite
Agents built on AWS Bedrock, Azure OpenAI and Vertex AI call models, MCP servers and internal APIs. Zenstra decides those calls at a gateway in your VPC; payloads never leave.
Homegrown agents are built to do real work: read accounts, update cases, move money. That makes every call they make a decision worth checking.
Bedrock agentA loan-servicing agent built on AWS Bedrock runs inside your VPC.grantIt asks for access to the loan system and receives a grant for 15 minutes.requestA crafted request asks it to transfer funds between accounts.mcp core-banking · accounts.transferPer-agent identity and short-lived grants for the models and tools each agent uses, decided at a gateway in your own cloud.
allowbedrock.invokefor loan-doc-classifier · expires 15mdenymcp.callwhen tool not in grantholdpayments.*for human approvaldenynet.egressto unapproved destinationsThe gateway runs where your agents run, in front of every model, MCP server and API.
Deploy the gateway with Helm or Terraform in AWS, Azure or Google Cloud, in your VPC or Kubernetes cluster.
Each agent gets its own identity and short-lived grants for the models and tools it uses.
Every call is decided inline, before it reaches the tool. If Zenstra is unreachable, actions are held.
Bedrock, Azure OpenAI and Vertex AI, per-agent credentials.
Each tool call checked against the agent’s grant.
Core, LOS and CRM connectors as scoped grants.
Held for a person to approve, by policy.
Approved destinations only.
Attested before actions run.
Decisions are made at 18 ms at p95, inline with the call.
Only decisions and redacted evidence. Prompts and payloads stay in your environment.
AWS, Azure and Google Cloud, deployed in your VPC or Kubernetes cluster.