Platform, security, and AI engineering teams selecting a production control architecture.
When to use an MCP gateway, an API gateway, or both.
The control gap
The names are often used interchangeably even though they answer different control questions. An API gateway governs conventional service traffic; an MCP gateway understands clients, tools, and invocation context presented through MCP.
An API gateway normally starts with an HTTP or service route and applies controls such as token validation, rate limits, network policy, and request transformation. An MCP gateway starts with an agent-facing protocol where a client discovers tools and invokes them with structured arguments. Tool identity, server identity, tool schemas, and agent delegation are therefore first-class decision inputs.
The products can be layered. The MCP gateway can decide whether an agent may call `create_refund` for a particular account, while the API gateway protects the downstream payments service and its conventional endpoints. Buying one and calling it both leaves a context gap: either agent semantics disappear before policy, or service controls are duplicated poorly at the MCP edge.
What good looks like
A purchase decision based on the enforcement point and evidence the team actually needs, rather than category labels.
- Protocol and discovery support
- Identity propagation
- Tool and argument awareness
- Approval before side effects
A production workflow
- Map where an MCP gateway sits in the request path and which failures it can stop.
- Repeat the exercise for an API gateway, including identities and resource context visible at decision time.
- Classify required actions as automatic, denied, or human-approved.
- Run representative calls in shadow mode and compare the evidence produced by each design.
Evidence to require
- Enforcement occurs before the external side effect, not after log ingestion.
- Decisions identify the agent and delegated user as separate principals.
- The resource, environment, policy version, reason, and response are retained together.
- Approval is bound to one request and expires instead of creating standing access.
Buyer checklist
- Which option can block the side effect at the moment of execution?
- Which identities and resource attributes are visible to its policy engine?
- Does it support request-bound human approval?
- Can the team export a complete decision trail without reconstructing several logs?
Practical answers
Common implementation questions
Is an MCP gateway a replacement for an API gateway?
Usually not. The right design depends on the enforcement point, protocol, and decision context. Many production systems use both, with each protecting the layer it can actually observe.
Where does Endram fit?
Endram is the runtime authorization and approval layer for agent tool calls. It evaluates the requested action before execution and keeps the decision evidence.
Can this be tested without interrupting production?
Yes. Shadow mode shows how candidate policies classify real calls before enforcement is enabled.