Add access tier documentation for Enhanced Experience

modules/ROOT/nav.adoc

@@ -6,7 +6,10 @@
     *** xref:exp-glossary.adoc[Glossary]
   ** xref:exp-release-notes.adoc[Release Notes]
   ** xref:exp-start-home.adoc[]
+    *** xref:exp-understand-access-tiers.adoc[]
     *** xref:exp-view-portfolio-overview.adoc[]
+    *** xref:exp-get-started-api-portfolio.adoc[]
+    *** xref:exp-get-started-agent-portfolio.adoc[]
   ** xref:exp-add-services-to-portfolio.adoc[]
     *** xref:exp-connect-providers-to-add-services.adoc[]
     *** xref:exp-register-services-manually.adoc[]

modules/ROOT/pages/exp-add-semantic-services.adoc

@@ -1,5 +1,10 @@
 = Adding Semantic Services
 
+[NOTE]
+====
+Semantic services require Agent + API Portfolio Access. If you don't see the Semantic Services option under LLM Proxies, contact your administrator to verify your subscription tier. See xref:exp-understand-access-tiers.adoc[] for more information.
+====
+
 Semantic services improve how the system interprets and routes LLM-driven requests across your portfolio. They help teams apply context-aware behavior so the system directs requests to the most relevant tools, services, or pathways that your organization configured.
 
 Use this capability when your workflows need more than static routing. With semantic context in place, the system can support better request matching, more consistent response behavior, and stronger control as usage grows across agents, APIs, MCP servers, and related integrations.

modules/ROOT/pages/exp-create-mcp-server.adoc

@@ -1,5 +1,10 @@
 = Create an MCP Server
 
+[NOTE]
+====
+MCP server management requires Agent + API Portfolio Access. If you don't see the MCP Servers catalog, contact your administrator to verify your subscription tier. See xref:exp-understand-access-tiers.adoc[] for more information.
+====
+
 Create MCP servers to define identity, runtime connectivity, and the tools and resources exposed to MCP clients.
 
 Creating an MCP server defines a new MCP implementation—for example, by transcoding an existing REST API or creating a runtime deployment—and establishes the server identity, runtime connectivity, and the MCP capabilities (tools and resources) exposed to clients.

modules/ROOT/pages/exp-get-started-agent-portfolio.adoc

@@ -0,0 +1,288 @@
+= Get Started with Agent Portfolio Management
+
+[NOTE]
+====
+Agent portfolio management requires Agent + API Portfolio Access. If you don't see the Agents, MCP Servers, or LLM Proxies catalogs, contact your administrator to verify your subscription tier. See xref:exp-understand-access-tiers.adoc[] for more information.
+====
+
+Use the enhanced experience to register, govern, monitor, and optimize your AI-powered services, including agents, MCP servers, and LLM proxies. The agent portfolio management features provide visibility and control over your AI service estate alongside traditional APIs.
+
+== Before You Begin
+
+Confirm these requirements with your administrator:
+
+Access::
+* You have Agent + API Portfolio Access.
+* You can sign in to the enhanced experience through your organization's entry point.
+
+Permissions::
+* You have the necessary Access Management permissions to view and manage agents, MCP servers, and LLM proxies.
+* See xref:exp-start-home.adoc#permissions[Enhanced Experience Permissions] for details.
+
+Integration::
+* Your organization has configured connections to Anypoint Platform.
+* If using external AI providers, your administrator has configured the appropriate provider connections under *Platform* > *Providers*.
+
+== Agent Portfolio Workflow
+
+Follow these steps to start managing your agent portfolio:
+
+=== 1. Navigate to the Agent Catalogs
+
+. Sign in to the enhanced experience.
+. In the main navigation, select *Portfolio*.
+. Review the available catalogs:
++
+* *Agents*: AI-powered services that use the Agent-to-Agent (A2A) protocol
+* *MCP Servers*: Services that expose tools and resources via the Model Context Protocol
+* *LLM Proxies*: Gateway-backed proxies to large language models
+* *APIs*: Traditional REST, GraphQL, and AsyncAPI services
+
+=== 2. Register Your Agent Services
+
+Add agents to the catalog using one of these methods:
+
+Manual Registration with A2A Endpoint::
+Use this method when your agent exposes an A2A endpoint:
++
+. On the Agents catalog page, select *Add Agent*.
+. Enter the A2A endpoint URL.
+. The system fetches the agent card automatically.
+. Review the agent metadata and capabilities.
+. Select *Save*.
+
+Manual Registration with Agent Card::
+Use this method when you have an agent card file:
++
+. On the Agents catalog page, select *Add Agent*.
+. Choose to upload an agent card file.
+. Complete the registration form with agent metadata.
+. Select *Save*.
+
+Provider Discovery::
+Use scanners to automatically discover agents from connected AI platforms:
++
+. Configure a provider connection under *Platform* > *Providers*.
+. Create a scanner for the provider.
+. Run the scanner to discover agents automatically.
+. Review and approve discovered agents.
+
+For detailed steps, see xref:exp-register-services-manually.adoc[] and xref:exp-connect-providers-to-add-services.adoc[].
+
+=== 3. Register MCP Servers
+
+Add MCP servers to the catalog:
+
+Create from Existing API::
+. Navigate to an API detail page.
+. Select *Actions* > *Create MCP Server*.
+. Configure the MCP server endpoint and schema.
+. Select *Create*.
+
+Manual Registration::
+. On the MCP Servers catalog page, select *Add MCP Server*.
+. Enter the MCP URL or upload an MCP schema file.
+. Complete the registration form.
+. Select *Save*.
+
+For detailed steps, see xref:exp-create-mcp-server.adoc[].
+
+=== 4. Register LLM Proxies
+
+Add LLM proxies to the catalog:
+
+. On the LLM Proxies catalog page, select *Add LLM Proxy*.
+. Select the target LLM provider and model.
+. Configure the proxy instance on a gateway.
+. Set up authentication and rate limiting policies.
+. Select *Create*.
+
+=== 5. View Service Details
+
+. From any catalog, select a service to open its detail page.
+. Review the available tabs:
++
+* *Overview*: Basic metadata, description, status, and capabilities
+* *Versions*: All registered versions (where applicable)
+* *Instances*: Deployed instances on gateways
+* *Policies*: Applied runtime policies
+* *Conformance Report*: Governance compliance status
+* *Monitoring*: Performance metrics, token usage, and health data
+
+For more information, see xref:exp-view-service-details.adoc[].
+
+=== 6. Create Instances
+
+Deploy your agent services to gateways:
+
+. On the service detail page, select the *Instances* tab.
+. Select *Add Instance*.
+. Choose the gateway type:
++
+* *Managed instance*: Routes traffic through Anypoint Omni Gateway with full policy enforcement, authentication, and monitoring
+* *Unmanaged instance*: Lighter deployment without Omni Gateway
+. Configure the instance URL and target endpoint.
+. Select *Create*.
+
+For detailed steps, see xref:exp-add-instances.adoc[].
+
+=== 7. Apply Governance Policies
+
+Enforce runtime controls on your agent service instances:
+
+Instance-Level Policies::
+. On the service detail page, select the *Instances* tab.
+. Select an instance.
+. On the instance detail page, select the *Policies* tab.
+. Select *Apply Policy*.
+. Choose a policy template appropriate for agent services:
++
+* Authentication and authorization
+* Rate limiting and throttling
+* Tool mapping (control which tools an agent can invoke)
+* Tool sanitization (filter or modify tool inputs/outputs)
+* Cost controls
+. Configure policy parameters.
+. Select *Apply*.
+
+Organization-Wide Strategies::
+Create governance strategies that automatically apply policies to matching agent services:
++
+. Navigate to *Governance* > *Governance Strategies*.
+. Create a new strategy with rules, scope, and automated policies.
+. Select domains relevant to AI services (performance, cost, security, compliance).
+. Activate the strategy.
++
+For detailed steps, see xref:exp-create-governance-strategies.adoc[].
+
+=== 8. Monitor Agent Performance
+
+Track runtime health and usage:
+
+. On the service detail page, select the *Monitoring* tab.
+. Review key metrics:
++
+* Request volume and throughput
+* Response times and latency
+* Token usage and consumption
+* Error rates and failure modes
+* Tool invocation patterns (for agents)
+. Set up alerts for threshold violations under *Observability* > *Notifications*.
+
+For more information, see xref:exp-monitor-your-services.adoc[] and xref:exp-view-detailed-metrics-for-services.adoc[].
+
+=== 9. Review Compliance
+
+Check governance conformance for agent services:
+
+. On the service detail page, select the *Conformance Report* tab.
+. Review the compliance score and severity breakdown.
+. Identify rule violations and warnings related to:
++
+* Agent card schema compliance
+* Security and authentication requirements
+* Tool exposure policies
+* Cost and performance thresholds
+. Address high-priority findings by updating agent configurations or applying policies.
+
+=== 10. Manage Costs
+
+Track token usage and optimize AI spending:
+
+. Navigate to *Governance* > *Cost Management*.
+. Review token usage by service, instance, and time period.
+. Identify cost optimization opportunities:
++
+* High-volume agents with potential for caching
+* Inefficient tool usage patterns
+* Opportunities for tool mapping to reduce unnecessary invocations
+. Apply cost reduction policies:
++
+* Tool mapping to limit agent tool access
+* Rate limiting to control request volume
+* Caching policies to reduce redundant LLM calls
+
+For more information, see xref:exp-view-cost-and-token-usage.adoc[].
+
+== Use Semantic Search
+
+Search across your agent portfolio using natural language:
+
+. Navigate to *Portfolio*.
+. Use the search bar at the top of any catalog.
+. Enter a natural language query describing what you're looking for.
+. Review semantically matched results across agents, MCP servers, and APIs.
+
+Semantic search helps you find relevant services based on capabilities and descriptions, not just exact name matches.
+
+== Next Steps
+
+After establishing your agent portfolio:
+
+* Set up automated scanners for AI provider discovery
+* Define agent-specific governance strategies
+* Configure cost thresholds and alerts
+* Integrate with Slack for agent notifications
+* Connect the Platform MCP Server to Claude Desktop for agent development workflows
+* Explore the Ask MuleSoft AI assistant for guided portfolio management
+
+== Common Tasks
+
+=== Update an Agent's Capabilities
+
+. Navigate to the agent detail page.
+. Select *Actions* > *Update Agent Card*.
+. Upload the updated agent card or point to the refreshed A2A endpoint.
+. Review changes and save.
+
+=== Configure Tool Mapping for an Agent
+
+. Navigate to the agent instance detail page.
+. Select the *Policies* tab.
+. Select *Apply Policy* > *Tool Mapping*.
+. Define which tools the agent is allowed to invoke.
+. Save the policy configuration.
+
+=== Review Token Usage Trends
+
+. Navigate to *Governance* > *Cost Management*.
+. Filter by agent or MCP server.
+. Review the usage timeline chart.
+. Export data for further analysis if needed.
+
+=== Create an MCP Server from an API
+
+. Navigate to an existing API detail page.
+. Select *Actions* > *Create MCP Server*.
+. Configure how API operations map to MCP tools.
+. Save the MCP server configuration.
+. The new MCP server appears in the MCP Servers catalog.
+
+For detailed steps, see xref:exp-create-mcp-server.adoc[].
+
+== Troubleshooting
+
+*I can't see the Agents, MCP Servers, or LLM Proxies catalogs*::
+Verify that your subscription includes Agent + API Portfolio Access. Contact your administrator to confirm your subscription tier and Access Management permissions. See xref:exp-understand-access-tiers.adoc[].
+
+*My agent registration fails*::
+Check that your A2A endpoint is accessible and returns a valid agent card. Review error messages for specific validation issues. Verify that your agent card conforms to the A2A protocol schema.
+
+*Token usage data is not appearing*::
+Confirm that your agent instance is deployed to a managed gateway and is receiving traffic. Token usage data may take a few minutes to appear after instance creation. Verify that the gateway is configured to report usage metrics.
+
+*Semantic search returns no results*::
+Semantic search requires indexed service descriptions. Newly registered services may take a few minutes to appear in semantic search results. Verify that your services have detailed descriptions in their metadata.
+
+*Tool mapping policies are not being enforced*::
+Verify that the agent instance is a managed instance on Anypoint Omni Gateway. Unmanaged instances do not support policy enforcement through the enhanced experience. Confirm that the tool mapping policy is applied and active.
+
+== See Also
+
+* xref:exp-understand-access-tiers.adoc[]
+* xref:exp-overview.adoc[]
+* xref:exp-register-services-manually.adoc[]
+* xref:exp-create-mcp-server.adoc[]
+* xref:exp-add-instances.adoc[]
+* xref:exp-work-with-governance-strategies.adoc[]
+* xref:exp-view-cost-and-token-usage.adoc[]