ReportPortal MCP Server

What is ReportPortal MCP Server?

The ReportPortal MCP Server bridges a ReportPortal instance with AI chat assistants (e.g., Claude Desktop, GitHub Copilot, Cursor), letting users ask plain-English questions about test runs and get answers directly from ReportPortal. Starting with version 2.x, it also integrates with the ReportPortal Test Case Management System (TMS) — test cases, folders, milestones, and test plans — so users can browse and manage test design data through the same chat interface.

How to use ReportPortal MCP Server?

Install the server locally via Docker (recommended) or a pre-built binary, then configure your AI tool’s MCP settings (e.g., mcp.json) with the required environment variables: RP_HOST, RP_API_TOKEN, and optionally RP_PROJECT. Alternatively, connect to a remotely deployed server in HTTP mode by providing its URL and authentication headers. The server runs in stdio mode by default for local tools; set MCP_MODE=http for remote access.

Key features of ReportPortal MCP Server

• Query test runs, failures, and error details via natural language
• Browse, create, and update TMS test cases, folders, milestones, and test plans
• Works with any MCP-compatible AI tool (Claude, Cursor, Copilot, etc.)
• No custom scripts needed — common queries are built-in as chat commands
• Supports local (stdio) and remote (HTTP/SSE) transport modes
• Docker image and pre-built binaries available for easy installation

Use cases of ReportPortal MCP Server

• QA tester asks “What tests failed in the last run?” and gets an instant summary
• Manager lists the 5 most recent test runs without navigating the UI
• User creates a test plan for the current sprint milestone via chat
• Team browses test cases in a specific folder and updates them through the AI assistant
• Developer retrieves error details from a failed build while staying in the editor

FAQ from ReportPortal MCP Server

What information do I need to set up the server?

You need the ReportPortal instance URL (RP_HOST), an API token (RP_API_TOKEN), and optionally the project key (RP_PROJECT). The project key is the unique project identifier from the ReportPortal settings page, not the display name.

How do I keep my API token secure?

Never commit tokens to version control or share them publicly. Store them in environment variables or a secure configuration file.

Can I use the server in a remote environment?

Yes. Set the system environment variable MCP_MODE=http to run the server in HTTP/SSE transport mode, then AI tools can connect to it via its remote URL with authentication headers.

What transport modes are supported?

The server supports stdio (default, for local integrations with Claude Desktop, Cursor, VS Code, etc.) and http (for remote or multi-client access). The mode is controlled by the MCP_MODE environment variable.

Does the server support the Test Case Management System (TMS)?

Yes, starting with version 2.x. You can manage test cases, folders, milestones, and test plans through the same chat interface — for example, “List test cases in the Smoke folder” or “Create a test plan for the current sprint milestone.”