	Omni Gateway	Mule Gateway
Summary	Envoy-based API gateway to secure all APIs, Mule and non-Mule, running anywhere	Java-based API gateway embedded into Mule, to secure only Mule APIs
Tech Stack	Underlying engine built upon Envoy Leverages Fluent Bit for logging	Java Spring Application embedded into Mule
Use Case	High performant and high availability Secure any API anywhere.	Secure a Mule API basic endpoint, or enable a dedicated proxy as an embedded library in a Mule instance. Also available as a Mule proxy application in CloudHub
Key Capabilities	Small footprint Multiple deployment patterns and modes, including as a native Kubernetes Ingress controller Connected Mode as well as Local Mode available	Same technology as Mule integration applications API Autodiscovery for Mule applications Connected Mode only
Custom Policy Development	Rust WASM SDKs	Java and Mule DSL

Omni Gateway

Mule Gateway

Summary

Envoy-based API gateway to secure all APIs, Mule and non-Mule, running anywhere

Java-based API gateway embedded into Mule, to secure only Mule APIs

Tech Stack

Underlying engine built upon Envoy
Leverages Fluent Bit for logging

Java Spring Application embedded into Mule

Use Case

High performant and high availability
Secure any API anywhere.

Secure a Mule API basic endpoint, or enable a dedicated proxy as an embedded library in a Mule instance.
Also available as a Mule proxy application in CloudHub

Key Capabilities

Small footprint
Multiple deployment patterns and modes, including as a native Kubernetes Ingress controller
Connected Mode as well as Local Mode available

Same technology as Mule integration applications
API Autodiscovery for Mule applications
Connected Mode only

Custom Policy Development

Rust WASM SDKs

Java and Mule DSL

Policy	Summary	Managed Omni Gateway	Omni Gateway Connected Mode	Omni Gateway Local Mode	Mule Gateway
A2A Agent Card	Rewrites the Agent card URL to match the server instance public URL	Yes	Yes	No	No
A2A PII Detector	Detects personally identifiable information (PII) in requests sent to the agent	Yes	Yes	No	No
A2A Prompt Decorator	Decorates prompts with context information	Yes	Yes	No	No
A2A Quality Evaluation	Evaluates completed A2A agent task responses using an LLM to score quality and completeness	Yes	Yes	No	No
A2A Schema Validation	Validates Agent requests to ensure they conform to the A2A specification	Yes	Yes	No	No
A2A Token Based Rate Limit	Limits API usage based on the number of GPT-4o-mini tokens consumed by the agents	Yes	Yes	No	No
A2A v1 Agent Card	Rewrites interface URLs in the A2A v1.0 agent card to match the gateway’s public address	Yes	Yes	No	No
A2A v1 Azure Content Guardrails	Evaluates A2A v1 agent messages against Azure AI Content Safety for harmful content, jailbreak attempts, hallucinations, and copyrighted material	Yes	Yes	No	No
A2A v1 Bedrock Guardrails	Evaluates A2A v1.0 agent requests and responses against Amazon Bedrock guardrails for content safety, PII detection, and contextual grounding	Yes	Yes	No	No
A2A v1 PII Detector	Detects personally identifiable information (PII) in A2A v1 requests	Yes	Yes	No	No
A2A v1 Prompt Decorator	Decorates prompts with context information on A2A v1 JSON-RPC and HTTP+JSON traffic	Yes	Yes	No	No
A2A v1 Schema Validation	Verifies that requests to the agent are compliant with the A2A v1.0 specification	Yes	Yes	No	No
A2A v1 to v0.3 Transcoder	Transcodes protocol messages bidirectionally between A2A v1.0 JSON-RPC clients and A2A v0.3 backend servers	Yes	Yes	No	No
A2A v1 Token Based Rate Limit	Rate limits v1 A2A agent usage of upstream resources	Yes	Yes	No	No
Agent Connection Telemetry	Tracks the number of requests made to agent or MCP server instances	Yes	Yes	No	No
API Security-Identification Header	Adds API instance identifiers as response headers for external security platform correlation	Yes	Yes	No	Yes
Amazon Bedrock Guardrails	Evaluates LLM prompts and responses against Amazon Bedrock guardrails for content safety, PII detection, and contextual grounding	Yes	Yes	No	No
Azure Content Safety	Evaluates LLM prompts and responses against Azure AI Content Safety for harmful content, jailbreak attempts, hallucinations, and copyrighted material	Yes	Yes	No	No
AWS Lambda	Allows access based on the basic authorization mechanism, with user-password defined on LDAP	Yes	Yes	Yes	No
AWS Request Signature	Signs outgoing requests with AWS Signature Version 4 (SigV4) or Signature Version 4A (SigV4A)	Yes	Yes	Yes	No
Basic Authentication: LDAP	Allows access based on the basic authorization mechanism, with user-password defined on LDAP	Yes	Yes	Yes	Yes
Basic Authentication: Simple	Allows access based on the basic authorization mechanism, with a single user-password	Yes	Yes	Yes	Yes
Client ID Enforcement	Allows access only to authorized client applications	Yes	Yes	No	Yes
Circuit Breaker	Prevents an upstream from receiving too many requests or connections	Yes	Yes	Yes	No
Credential Injection Basic Authentication	Injects a basic authorization header into outgoing requests	Yes	Yes	Yes	No
Credential Injection API Key	Injects an API key or client credentials headers into outgoing requests	Yes	Yes	Yes	No
Credential Injection OAuth 2.0	Injects an OAuth 2.0 authorization header into outgoing requests	Yes	Yes	Yes	No
Credential Injection JWT Generation	Generates and injects a JWT token into outgoing requests	Yes	Yes	Yes	No
OAuth 2.0 OBO Credential Injection	Exchanges incoming bearer tokens using OAuth 2.0 Token Exchange (RFC 8693), Microsoft Entra ID On-Behalf-Of, or OAuth 2.0 Token Exchange with CIBA	Yes	Yes	No	No
Cross-Origin Resource Sharing (CORS)	Enables access to resources residing in external domains	Yes	Yes	Yes	Yes
CrowdStrike AIDR Guardrails	Evaluates OpenAI-format LLM prompts and responses against CrowdStrike Falcon AI Detection and Response (AIDR), rejecting or redacting flagged content	Yes	Yes	No	No
DataWeave Body Transformation	Transforms request or response body content with a DataWeave script	Yes	Yes	Yes	No
DataWeave Headers Transformation	Transforms request or response headers with a DataWeave script	Yes	Yes	Yes	No
DataWeave Request Filter	Filters requests by using a DataWeave script	Yes	Yes	Yes	No
Detokenization	Returns a tokenized value to its original value	No	No	No	Yes
External Authorization	Authenticates requests by using an external gRPC or HTTP authorization service	Yes	Yes	Yes	No
External Processing	Sends the incoming HTTP requests or outgoing HTTP responses to an external gRPC service for additional processing	Yes	Yes	Yes	No
F5 Guardrails	Evaluates LLM prompts and responses against F5 AI Guardrails, blocking, flagging, or redacting flagged content	Yes	Yes	No	No
Google Model Armor Guardrails	Evaluates OpenAI and Gemini format LLM prompts and responses against a Google Cloud Model Armor template, blocking or redacting flagged content	Yes	Yes	No	No
GraphQL Introspection Control	Blocks or allows requests that access the `schema`, `type`, and `__typename` GraphQL meta fields	Yes	Yes	Yes	No
GraphQL Operation Limits	Limits GraphQL operation depth, aliases, root fields, and directives	Yes	Yes	Yes	No
GraphQL Schema Validation	Validates incoming GraphQL operations against a GraphQL schema definition	Yes	Yes	Yes	No
GraphQL Static Query Complexity	Calculates query complexity and blocks requests that exceed the defined budget	Yes	Yes	Yes	No
Header Injection	Adds headers to a request or a response	Yes	Yes	Yes	Yes
Header Removal	Removes headers from a request or a response	Yes	Yes	Yes	Yes
Health Check	Monitors API upstream health at specific intervals	Yes	Yes	No	No
HTTP Caching	Caches HTTP responses from an API implementation	Yes	Yes	Yes	Yes
Injection Protection	Scans request headers, path, query, and body to protect against SQL, XSS, and custom injection attacks in incoming requests	Yes	Yes	Yes	No
IP Allowlist	Allows a list or range of specified IP addresses to request access	Yes	Yes	Yes	Yes
IP Blocklist	Blocks a single IP address or a range of IP addresses from accessing an API endpoint	Yes	Yes	Yes	Yes
A2A In-Task Authorization Code	Ensures secondary credentials are present for in-task authentication using OAuth 2.0 Authorization Code flow	Yes	Yes	No	No
JavaScript Scripting	Runs a user-supplied JavaScript module to inspect and modify requests and responses	Yes	Yes	Yes	No
JSON Threat Protection	Protects against malicious JSON in API requests	Yes	Yes	Yes	Yes
JWT Validation	Validates a JWT	Yes	Yes	Yes	Yes
LLM PII Detection	Detects personally identifiable information (PII) in OpenAI and Anthropic format LLM requests and responses	Yes	Yes	No	No
LLM Token Based Rate Limit	Rate limits model proxy usage based on token consumption	Yes	Yes	No	No
MCP Attribute-Based Access Control	Controls access to tools, resources, and prompts based on user information such as Tiers, IP, Headers, or Claims	Yes	Yes	No	No
MCP Global Access	Restricts which MCP tools are exposed by defining Allow and Block rules	Yes	Yes	No	No
MCP PII Detector	Detects personally identifiable information (PII) in Model Context Protocol (MCP) JSON-RPC traffic on both the request and response sides	Yes	Yes	No	No
MCP Payload Optimization	Reduces tool response token consumption using one or more configurable methods: remove structural noise, trim oversized responses to match agent intent, or compress repeated structures into TOON	Yes	Yes	No	No
MCP Schema Validation	Validates MCP requests to ensure they conform to the MCP specification	Yes	Yes	No	No
MCP Support	Adds MCP support to an Omni Gateway MCP server instance. This policy is required for your MCP server instance to function properly.	Yes	Yes	No	No
MCP Tool Mapping	Redefines tool names and descriptions to reduce prompt complexity and improve LLM comprehension	Yes	Yes	No	No
MCP Tools Progressive Disclosure	Reduces context window token cost by exposing tools on-demand through search rather than listing all tools up front	Yes	Yes	No	No
Message Logging	Logs custom messages using information from incoming requests, responses from the backend, or information from other policies applied to the same API endpoint	Yes	Yes	Yes	Yes
Model Proxy Request Compression	Compresses message content and replayed tool results in Model Proxy request bodies before they reach the upstream provider, reducing billed tokens	Yes	Yes	No	No
Outbound Message Logging	Logs custom messages from outbound requests sent to upstream services, responses from upstream services, or information from other outbound policies applied to the same upstream service	Yes	Yes	Yes	No
OAuth 2.0 Access Token Enforcement Using Mule OAuth Provider	Allows access only to authorized client applications	No	No	No	Yes
OAuth 2.0 Protected Resource Metadata	Serves OAuth 2.0 Protected Resource Metadata at the well-known endpoint	Yes	Yes	No	No
OAuth 2.0 Token Introspection	Allows access only to authorized client applications	Yes	Yes	Yes	No
OpenAM OAuth 2.0 Token Enforcement	Allows access only to authorized client applications	No	No	No	Yes
OpenID Connect OAuth 2.0 Access Token Enforcement	Allows access only to authorized client applications	Yes	Yes	No	Yes
PingFederate OAuth 2.0 Token Enforcement	Allows access only to authorized client applications	No	No	No	Yes
Rate Limiting	Monitors access to an API by defining the maximum number of requests processed within a period of time	Yes	Yes	Yes	Yes
Rate Limiting: SLA-based	Monitors access to an API by defining the maximum number of requests processed within a timespan, based on SLAs	Yes	Yes	No	Yes
Regex Prompt Guard	Blocks LLM requests that match deny-list regex patterns	Yes	Yes	No	No
Response Timeout	Sets the maximum duration Omni Gateway waits for a response from an upstream service	Yes	Yes	Yes	No
Schema Validation	Validates incoming traffic against a supplied OAS3 schema	Yes	Yes	Yes	No
SOAP Schema Validation	Validates incoming traffic against a specified WSDL schema	Yes	Yes	Yes	No
SSE Logging	Logs every SSE event while streaming	Yes	Yes	No	No
Stream Idle Timeout	Sets the maximum duration a stream can remain idle without receiving a request or response	Yes	Yes	Yes	No
Spike Control	Regulates API traffic	Yes	Yes	Yes	Yes
Traffic Management for Multiple Upstream Services	Manages API instance traffic to multiple upstream services from a single consumer endpoint	Yes	Yes	Yes	No
Traffic Management for Multiple Upstream Services (Weighted)	Manages API instance traffic to multiple upstream services from a single consumer endpoint, using weighted percentages	Yes	Yes	Yes	No
Transport Layer Security (TLS) - Inbound	Enables authentication between a client and the API proxy	Yes	Yes	Yes	No
Transport Layer Security (TLS) - Outbound	Enables two-way authentication between the API proxy and an upstream service	Yes	Yes	Yes	No
Tokenization	Transforms sensitive data into a nonsensitive equivalent, named token	No	No	No	Yes
Upstream Idle Timeout	Sets the maximum duration a stream between Omni Gateway and an upstream service can remain idle	Yes	Yes	Yes	No
WebSocket Connection Limit	Limits the number of simultaneous WebSocket connections that each API instance allows	Yes	Yes	Yes	No
WS-Security	Validates Web Services Security (WS-Security) UsernameToken credentials, XML signatures, or both on inbound SOAP requests	Yes	Yes	Yes	No
XML Threat Protection	Protects against malicious XML in API requests	Yes	Yes	Yes	Yes

Policy

Summary

Managed Omni Gateway

Omni Gateway Connected Mode

Omni Gateway Local Mode

Mule Gateway

A2A Agent Card

Rewrites the Agent card URL to match the server instance public URL

Yes

A2A PII Detector

Detects personally identifiable information (PII) in requests sent to the agent

Yes

A2A Prompt Decorator

Decorates prompts with context information

Yes

A2A Quality Evaluation

Evaluates completed A2A agent task responses using an LLM to score quality and completeness

Yes

A2A Schema Validation

Validates Agent requests to ensure they conform to the A2A specification

Yes

A2A Token Based Rate Limit

Limits API usage based on the number of GPT-4o-mini tokens consumed by the agents

Yes

A2A v1 Agent Card

Rewrites interface URLs in the A2A v1.0 agent card to match the gateway’s public address

Yes

A2A v1 Azure Content Guardrails

Evaluates A2A v1 agent messages against Azure AI Content Safety for harmful content, jailbreak attempts, hallucinations, and copyrighted material

Yes

A2A v1 Bedrock Guardrails

Evaluates A2A v1.0 agent requests and responses against Amazon Bedrock guardrails for content safety, PII detection, and contextual grounding

Yes

A2A v1 PII Detector

Detects personally identifiable information (PII) in A2A v1 requests

Yes

A2A v1 Prompt Decorator

Decorates prompts with context information on A2A v1 JSON-RPC and HTTP+JSON traffic

Yes

A2A v1 Schema Validation

Verifies that requests to the agent are compliant with the A2A v1.0 specification

Yes

A2A v1 to v0.3 Transcoder

Transcodes protocol messages bidirectionally between A2A v1.0 JSON-RPC clients and A2A v0.3 backend servers

Yes

A2A v1 Token Based Rate Limit

Rate limits v1 A2A agent usage of upstream resources

Yes

Agent Connection Telemetry

Tracks the number of requests made to agent or MCP server instances

Yes

API Security-Identification Header

Adds API instance identifiers as response headers for external security platform correlation

Yes

Amazon Bedrock Guardrails

Evaluates LLM prompts and responses against Amazon Bedrock guardrails for content safety, PII detection, and contextual grounding

Yes

Azure Content Safety

Evaluates LLM prompts and responses against Azure AI Content Safety for harmful content, jailbreak attempts, hallucinations, and copyrighted material

Yes

AWS Lambda

Allows access based on the basic authorization mechanism, with user-password defined on LDAP

Yes

AWS Request Signature

Signs outgoing requests with AWS Signature Version 4 (SigV4) or Signature Version 4A (SigV4A)

Yes

Basic Authentication: LDAP

Allows access based on the basic authorization mechanism, with user-password defined on LDAP

Yes

Basic Authentication: Simple

Allows access based on the basic authorization mechanism, with a single user-password

Yes

Client ID Enforcement

Allows access only to authorized client applications

Yes

Circuit Breaker

Prevents an upstream from receiving too many requests or connections

Yes

Credential Injection Basic Authentication

Injects a basic authorization header into outgoing requests

Yes

Credential Injection API Key

Injects an API key or client credentials headers into outgoing requests

Yes

Credential Injection OAuth 2.0

Injects an OAuth 2.0 authorization header into outgoing requests

Yes

Credential Injection JWT Generation

Generates and injects a JWT token into outgoing requests

Yes

OAuth 2.0 OBO Credential Injection

Exchanges incoming bearer tokens using OAuth 2.0 Token Exchange (RFC 8693), Microsoft Entra ID On-Behalf-Of, or OAuth 2.0 Token Exchange with CIBA

Yes

Cross-Origin Resource Sharing (CORS)

Enables access to resources residing in external domains

Yes

CrowdStrike AIDR Guardrails

Evaluates OpenAI-format LLM prompts and responses against CrowdStrike Falcon AI Detection and Response (AIDR), rejecting or redacting flagged content

Yes

DataWeave Body Transformation

Transforms request or response body content with a DataWeave script

Yes

DataWeave Headers Transformation

Transforms request or response headers with a DataWeave script

Yes

DataWeave Request Filter

Filters requests by using a DataWeave script

Yes

Detokenization

Returns a tokenized value to its original value

Yes

External Authorization

Authenticates requests by using an external gRPC or HTTP authorization service

Yes

External Processing

Sends the incoming HTTP requests or outgoing HTTP responses to an external gRPC service for additional processing

Yes

F5 Guardrails

Evaluates LLM prompts and responses against F5 AI Guardrails, blocking, flagging, or redacting flagged content

Yes

Google Model Armor Guardrails

Evaluates OpenAI and Gemini format LLM prompts and responses against a Google Cloud Model Armor template, blocking or redacting flagged content

Yes

GraphQL Introspection Control

Blocks or allows requests that access the schema, type, and __typename GraphQL meta fields

Yes

GraphQL Operation Limits

Limits GraphQL operation depth, aliases, root fields, and directives

Yes

GraphQL Schema Validation

Validates incoming GraphQL operations against a GraphQL schema definition

Yes

GraphQL Static Query Complexity

Calculates query complexity and blocks requests that exceed the defined budget

Yes

Header Injection

Adds headers to a request or a response

Yes

Header Removal

Removes headers from a request or a response

Yes

Health Check

Monitors API upstream health at specific intervals

Yes

HTTP Caching

Caches HTTP responses from an API implementation

Yes

Injection Protection

Scans request headers, path, query, and body to protect against SQL, XSS, and custom injection attacks in incoming requests

Yes

IP Allowlist

Allows a list or range of specified IP addresses to request access

Yes

IP Blocklist

Blocks a single IP address or a range of IP addresses from accessing an API endpoint

Yes

A2A In-Task Authorization Code

Ensures secondary credentials are present for in-task authentication using OAuth 2.0 Authorization Code flow

Yes

JavaScript Scripting

Runs a user-supplied JavaScript module to inspect and modify requests and responses

Yes

JSON Threat Protection

Protects against malicious JSON in API requests

Yes

JWT Validation

Validates a JWT

Yes

LLM PII Detection

Detects personally identifiable information (PII) in OpenAI and Anthropic format LLM requests and responses

Yes

LLM Token Based Rate Limit

Rate limits model proxy usage based on token consumption

Yes

MCP Attribute-Based Access Control

Controls access to tools, resources, and prompts based on user information such as Tiers, IP, Headers, or Claims

Yes

MCP Global Access

Restricts which MCP tools are exposed by defining Allow and Block rules

Yes

MCP PII Detector

Detects personally identifiable information (PII) in Model Context Protocol (MCP) JSON-RPC traffic on both the request and response sides

Yes

MCP Payload Optimization

Reduces tool response token consumption using one or more configurable methods: remove structural noise, trim oversized responses to match agent intent, or compress repeated structures into TOON

Yes

MCP Schema Validation

Validates MCP requests to ensure they conform to the MCP specification

Yes

MCP Support

Adds MCP support to an Omni Gateway MCP server instance. This policy is required for your MCP server instance to function properly.

Yes

MCP Tool Mapping

Redefines tool names and descriptions to reduce prompt complexity and improve LLM comprehension

Yes

MCP Tools Progressive Disclosure

Reduces context window token cost by exposing tools on-demand through search rather than listing all tools up front

Yes

Message Logging

Logs custom messages using information from incoming requests, responses from the backend, or information from other policies applied to the same API endpoint

Yes

Model Proxy Request Compression

Compresses message content and replayed tool results in Model Proxy request bodies before they reach the upstream provider, reducing billed tokens

Yes

Outbound Message Logging

Logs custom messages from outbound requests sent to upstream services, responses from upstream services, or information from other outbound policies applied to the same upstream service

Yes

OAuth 2.0 Access Token Enforcement Using Mule OAuth Provider

Allows access only to authorized client applications

Yes

OAuth 2.0 Protected Resource Metadata

Serves OAuth 2.0 Protected Resource Metadata at the well-known endpoint

Yes

OAuth 2.0 Token Introspection

Allows access only to authorized client applications

Yes

OpenAM OAuth 2.0 Token Enforcement

Allows access only to authorized client applications

Yes

OpenID Connect OAuth 2.0 Access Token Enforcement

Allows access only to authorized client applications

Yes

PingFederate OAuth 2.0 Token Enforcement

Allows access only to authorized client applications

Yes

Rate Limiting

Monitors access to an API by defining the maximum number of requests processed within a period of time

Yes

Rate Limiting: SLA-based

Monitors access to an API by defining the maximum number of requests processed within a timespan, based on SLAs

Yes

Regex Prompt Guard

Blocks LLM requests that match deny-list regex patterns

Yes

Response Timeout

Sets the maximum duration Omni Gateway waits for a response from an upstream service

Yes

Schema Validation

Validates incoming traffic against a supplied OAS3 schema

Yes

SOAP Schema Validation

Validates incoming traffic against a specified WSDL schema

Yes

SSE Logging

Logs every SSE event while streaming

Yes

Stream Idle Timeout

Sets the maximum duration a stream can remain idle without receiving a request or response

Yes

Spike Control

Regulates API traffic

Yes

Traffic Management for Multiple Upstream Services

Manages API instance traffic to multiple upstream services from a single consumer endpoint

Yes

Traffic Management for Multiple Upstream Services (Weighted)

Manages API instance traffic to multiple upstream services from a single consumer endpoint, using weighted percentages

Yes

Transport Layer Security (TLS) - Inbound

Enables authentication between a client and the API proxy

Yes

Transport Layer Security (TLS) - Outbound

Enables two-way authentication between the API proxy and an upstream service

Yes

Tokenization

Transforms sensitive data into a nonsensitive equivalent, named token

Yes

Upstream Idle Timeout

Sets the maximum duration a stream between Omni Gateway and an upstream service can remain idle

Yes

WebSocket Connection Limit

Limits the number of simultaneous WebSocket connections that each API instance allows

Yes

WS-Security

Validates Web Services Security (WS-Security) UsernameToken credentials, XML signatures, or both on inbound SOAP requests

Yes

XML Threat Protection

Protects against malicious XML in API requests

Yes