Skip to content

mcp.md

Latest commit

 

History

History
91 lines (56 loc) · 4.11 KB

mcp.md

File metadata and controls

91 lines (56 loc) · 4.11 KB

tssc: Model Context Protocol Server (MCP)

Usage

Configure your agentic LLM client to run the following command, and use STDIO for communication with the server.

tssc mcp-server --image="ghcr.io/redhat-appstudio/tssc:latest"

Note: if you use Cursor, you can use the Cursor MCP configuration

Then use the recommended prompt to get started, and follow the instructions.

Instructions

When the MCP server is initialized, it will provide instructions for the LLM client to follow. The instructions are documented here.

The instructions provide a clear overview of what the MCP server is designed to do, how it works, and the expected sequence of tools to follow.

Prompt

This is the recommended prompt to instruct the AI agent to start a deployment.

You are a senior Red Hat OpenShift Platform Engineer. You leverage a specialized OpenShift-based platform focused on Trusted Software Supply Chain (TSSC), primarily consisting of:

- Advanced Cluster Security (ACS)
- OpenShift GitOps
- Trusted Artifact Signer (TAS)
- OpenShift Pipelines
- Trusted Profile Analyzer (TPA)
- Red Hat Developer Hub (DH)
- Quay and other container registries

Your primary function is to provide actionable commands: YAML manifests and `oc`/`kubectl` CLI commands, all tailored for OpenShift 4.17+. You prioritize security-hardened configurations and adherence to CNCF best practices.

Your interaction schema is built around Model Context Protocol (MCP) tools, enabling you to interact with various OpenShift work as a platform engineer. Your responses are structured to include a quick command block for immediate use.

Now let's get started on deploying TSSC.

Tools

The following MCP tools are exposed by the MCP server.

Configuration

This section covers for the features in tssc config subcommand.

tssc_config_get

  • Description: Get the existing TSSC configuration in the cluster, or return the default if none exists yet. Use the default configuration as the reference to create a new TSSC configuration for the cluster.
  • Arguments: None.

tssc_config_create

  • Description: Create a new TSSC configuration in the cluster, in case none exists yet. Use the defaults as the reference to create a new TSSC cluster configuration.
  • Arguments:
    • namespace (string):
      • Description: The main namespace for TSSC (.tssc.namespace), where Red Hat Developer Hub (DH) and other fundamental services will be deployed.
      • Default: "tssc".
    • setting (object):
      • Description: The global settings object for TSSC (.tssc.settings{}). When empty the default settings will be used.

Integration

Integrations with external services are managed via the tssc integration <integration-name> command. Since these integrations often require sensitive information, credentials, the MCP server will not handle them directly. Instead, it will provide the user with the exact tssc integration commands needed to configure the integration.

tssc_integration_list

  • Description: List the TSSC integrations available for the user. Certain integrations are required for certain features, make sure to configure the integrations accordingly.
  • Arguments: None.

Deploy

To handle the long-running deployment process without blocking the MCP server, the MCP server will delegate tasks to the cluster.

The tssc design maintains a single deployment per cluster by centralizing the installer configuration in a unique ConfigMap, ensuring a single RHADS installation per cluster.

Following the same principle, the MCP server generates a Kubernetes Job to run the tssc container image (specifically the tssc deploy subcommand) which proceeds with installing the predefined sequence of Helm charts that makes up RHADS.

tssc_deploy_status

  • Description: Reports the status of the TSSC deploy Job running in the cluster.
  • Arguments: None.

tssc_deploy

  • Description: Deploys TSSC components to the cluster, uses the cluster configuration to deploy the TSSC components sequentially.
  • Arguments: None.