Overview
Zscaler MCP Server enables organizations to achieve:
Streamlined operations through natural language that empower teams to configure, monitor, and manage services more intuitively. Everyday requests are translated into API SDK calls, reducing complexity, minimizing human error, and improving efficiency.
Unified and secure access to APIs by providing a consistent, governed interface for interacting with services. This ensures least-privileged use of API calls, eliminates fragmented tools, and enforces strong governance across environments.
Accelerated innovation and automation at scale through rapid integration of new workflows and pipelines. MCP Server shortens time-to-value, simplifies orchestration, and helps enterprises scale automation with confidence.
Broad applicability across IT and DevOps, from configuration and monitoring to orchestration. MCP Server supports the full Zscaler portfolio - including Zscaler Internet Access, Zscaler Private Access, Zscaler Digital Experience, Zscaler Cloud and Branch Connector, Zscaler Client Connector, and Zscaler ZIdentity. MCP Server is offered as a bring-your-own-license (BYOL) solution, requiring existing entitlements to Zscaler services.
Highlights
- Simplify complex operations - Translate natural language into API calls, making configuration, monitoring, and troubleshooting faster and more intuitive for admins and developers.
- Unify and secure API access - Provide a single, consistent way to interact with your services, reducing complexity while maintaining strong security and governance.
- Enable faster innovation at scale - Accelerate integration of new workflows and automation pipelines, reducing time-to-value and supporting enterprise-wide agility.
Details
Unlock automation with AI agent solutions

Features and programs
Financing for AWS Marketplace purchases
Pricing
Vendor refund policy
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Amazon Bedrock AgentCore
- Amazon Bedrock AgentCore - Preview
Container image
Containers are lightweight, portable execution environments that wrap server application software in a filesystem that includes everything it needs to run. Container applications run on supported container runtimes and orchestration services, such as Amazon Elastic Container Service (Amazon ECS) or Amazon Elastic Kubernetes Service (Amazon EKS). Both eliminate the need for you to install and operate your own container orchestration software by managing and scheduling containers on a scalable cluster of virtual machines.
Version release notes
Initial release of Zscaler MCP Server for AWS AgentCore
Additional details
Usage instructions
Amazon Bedrock AgentCore Deployment (Zscaler MCP Server)
Prerequisites
- AgentCore enabled in your AWS account.
- Active subscription to ZIA or ZPA.
- Outbound HTTPS to Zscaler APIs. AgentCore provides this by default; optionally attach to a VPC so egress can be inspected/secured.
- Zscaler OneAPI credentials set as environment variables during hosting. Generate in ZIdentity as outlined in the README:
https://github.com/zscaler/zscaler-mcp-server?tab=readme-ov-file#zscaler-api-credentials--authenticationÂ- ZSCALER_CLIENT_ID
- ZSCALER_CLIENT_SECRET
- ZSCALER_VANITY_DOMAIN
- ZSCALER_CUSTOMER_ID (required for ZPA)
- ZSCALER_CLOUD (optional; set to beta only when using the Beta Tenant)
- Legacy API auth is also supported. See full guide (linked at the bottom) for variables and usage.
Step 1 - Create the IAM execution role (Console)
- IAM -> Roles -> Create role.
- Attach a custom policy or equivalent managed policies with these permissions:
-
Amazon ECR (image pull):
ecr:BatchGetImage, ecr:GetDownloadUrlForLayer, ecr:GetAuthorizationToken
Alternative: AmazonEC2ContainerRegistryReadOnly managed policy. -
CloudWatch Logs (write):
logs:DescribeLogStreams, logs:CreateLogGroup, logs:DescribeLogGroups, logs:CreateLogStream, logs:PutLogEvents
Alternative: CloudWatchFullAccess (or a scoped equivalent). -
AWS X-Ray (telemetry):
xray:PutTraceSegments, xray:PutTelemetryRecords, xray:GetSamplingRules, xray:GetSamplingTargets
Alternative: AWSXRayDaemonWriteAccess. -
CloudWatch metrics:
cloudwatch:PutMetricData (optionally restrict namespace to bedrock-agentcore). -
AgentCore access tokens:
bedrock-agentcore:GetWorkloadAccessToken (optionally scope to your default workload identity directory and the identities for this agent). -
Bedrock model invocation:
bedrock:InvokeModel, bedrock:InvokeModelWithResponseStream (optionally scope to required foundation models and your resources).
- Trust relationship: service principal bedrock-agentcore.amazonaws.com.
(Optionally add conditions restricting aws:SourceAccount to your account ID and aws:SourceArn to your AgentCore ARNs.) - Save and copy the Role ARN (you will provide this when hosting the agent).
Step 2 - Host the Agent (Console)
- Open Amazon Bedrock AgentCore -> Agent Runtimes -> Host Agent.
- Enter a Name (e.g., zscaler-mcp) and Description.
- Container image URI: paste the exact URI+tag shown on this Marketplace version (example):
709825985650.dkr.ecr.us-east-1.amazonaws.com/zscaler/zscaler-mcp-server:0.2.1-bedrock
Do not use latest. - Execution role: select the role created in Step 1.
- Protocol: MCP.
- Inbound identity: choose IAM (typical) or JWT if your environment requires JWT pass-through.
- Environment variables (add each key/value):
ZSCALER_CLIENT_ID, ZSCALER_CLIENT_SECRET, ZSCALER_VANITY_DOMAIN, ZSCALER_CUSTOMER_ID (if ZPA), ZSCALER_CLOUD (optional). - Click Host Agent and wait for Status: Active.
Step 3 - Verify
- Confirm the runtime is Active.
- Check CloudWatch Logs under /aws/bedrock-agentcore/runtimes/... for initialization and invocation details.
- From the AgentCore sandbox, invoke a simple call payload such as:
- List tools {"jsonrpc":"2.0","id":1,"method":"tools/list","params":{"_meta":{"progressToken":1}}}
- List DLP Dictionaries {"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"zia_dlp_dictionaries","arguments":{}}}
See the full guide for examples and troubleshooting tips.
Full documentation: Zscaler MCP Server - Amazon Bedrock AgentCore Guide
https://zscaler-mcp-server.readthedocs.io/en/latest/guides/amazon-bedrock-agentcore.htmlÂ
Support
Vendor support
Yes
Zscaler global support is available around the clock, with dedicated customer support engineers providing personalized assistance to ensure that customers are getting the most value from our products. Our support engineers have significant experience in networking and security, working closely with operations, sales, and engineering teams to ensure rapid response and resolution.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
