mcp-rp

mcp-rp

Enables AI assistants to interact with ReportPortal, covering launches, test items, logs, dashboards, widgets, filters, projects, users, and integrations.

Category
访问服务器

README

MCP Server for ReportPortal

An MCP (Model Context Protocol) server that exposes ReportPortal API operations as tools for AI assistants. Covers launches, test items, logs, dashboards, widgets, filters, projects, users, and integrations.

Prerequisites

  • Python 3.11+
  • A running ReportPortal instance (5.x)
  • An API token or username/password credentials

Installation

pip install -r requirements.txt

Configuration

The server reads configuration from three sources (in priority order):

  1. Environment variables (RP_URL, RP_TOKEN, RP_PROJECT, etc.)
  2. .env file in the project root
  3. config.yaml file

Quick Start

Copy the example files and fill in your values:

cp .env.example .env
# Edit .env with your ReportPortal URL, token, and project

Or use a YAML config:

cp config.example.yaml config.yaml
# Edit config.yaml with your settings

Environment Variables

Variable Description Required
RP_URL ReportPortal server URL Yes
RP_TOKEN API token (from Profile > API Keys) Yes*
RP_PROJECT Project name Yes
RP_USERNAME Username (alternative to token) No*
RP_PASSWORD Password (alternative to token) No*
RP_VERIFY_SSL Verify SSL certificates (default true) No

* Either RP_TOKEN or both RP_USERNAME and RP_PASSWORD are required.

Usage

Run with stdio transport (for Cursor, Claude Desktop, etc.)

python main.py

Run with a specific config file

python main.py /path/to/config.yaml

Cursor MCP Configuration

Add to your .cursor/mcp.json:

{
  "mcpServers": {
    "reportportal": {
      "command": "python",
      "args": ["/path/to/mcp-rp/main.py"],
      "env": {
        "RP_URL": "https://reportportal.example.com",
        "RP_TOKEN": "your-api-token",
        "RP_PROJECT": "my_project"
      }
    }
  }
}

Available Tools

Launches (10 tools)

Tool Description
rp_list_launches List launches with filters (status, name, mode, pagination)
rp_get_launch Get launch by numeric ID
rp_get_launch_by_uuid Get launch by UUID
rp_create_launch Start a new launch
rp_finish_launch Finish a running launch
rp_force_finish_launch Force-finish a stuck launch
rp_update_launch Update launch metadata
rp_delete_launch Delete a launch
rp_merge_launches Merge multiple launches
rp_compare_launches Compare launch statistics

Test Items (11 tools)

Tool Description
rp_list_test_items List test items with filters
rp_get_test_item Get test item by ID
rp_get_test_item_by_uuid Get test item by UUID
rp_create_test_item Create a suite, test, or step
rp_finish_test_item Finish a test item with status
rp_update_test_item Update test item metadata
rp_delete_test_item Delete a test item
rp_get_test_item_history Get execution history
rp_update_test_item_issues Bulk-update defect types
rp_link_external_issue Link bug-tracker tickets
rp_unlink_external_issue Unlink bug-tracker tickets

Logs (7 tools)

Tool Description
rp_list_logs List logs with filters
rp_get_log Get log by ID
rp_get_log_by_uuid Get log by UUID
rp_create_log Create a log entry
rp_get_nested_step_logs Get logs for nested steps
rp_search_logs Search for similar logs
rp_delete_log Delete a log entry

Dashboards (7 tools)

Tool Description
rp_list_dashboards List all dashboards
rp_get_dashboard Get dashboard with widgets
rp_create_dashboard Create a new dashboard
rp_update_dashboard Update dashboard metadata
rp_delete_dashboard Delete a dashboard
rp_add_widget_to_dashboard Add a widget to a dashboard
rp_remove_widget_from_dashboard Remove a widget

Widgets (5 tools)

Tool Description
rp_list_widget_names List all widget names
rp_get_widget Get widget details
rp_create_widget Create a new widget
rp_update_widget Update widget configuration
rp_preview_widget Preview widget data

Filters (5 tools)

Tool Description
rp_list_filters List all user filters
rp_get_filter Get filter details
rp_create_filter Create a new filter
rp_update_filter Update a filter
rp_delete_filter Delete a filter

Projects (8 tools)

Tool Description
rp_list_projects List all projects
rp_get_project Get project details
rp_get_project_names Get lightweight project name list
rp_create_project Create a new project
rp_update_project Update project configuration
rp_delete_project Delete a project
rp_assign_user_to_project Assign a user with a role
rp_unassign_user_from_project Remove a user

Users (6 tools)

Tool Description
rp_get_current_user Get authenticated user profile
rp_get_user Get user by login name
rp_search_users Search users by name/email
rp_list_api_keys List API keys for a user
rp_create_api_key Create a new API key
rp_delete_api_key Revoke an API key

Integrations (8 tools)

Tool Description
rp_list_global_integrations List global integrations
rp_list_project_integrations List project integrations
rp_get_integration Get integration details
rp_create_global_integration Create a global integration
rp_create_project_integration Create a project integration
rp_update_integration Update an integration
rp_delete_integration Delete an integration
rp_test_integration_connection Test integration connectivity

Architecture

main.py                  FastMCP server entry point
src/
  config/settings.py     Pydantic settings, YAML + env var loading
  clients/
    base.py              Abstract HTTP client interface
    reportportal.py      aiohttp client with OAuth, retries, rate limiting
  models/                Pydantic models for all RP entities
  services/              One service per domain (launch, test_item, log, ...)
  tools/                 MCP tool registrations per domain
  utils/
    logging.py           Structured logging (structlog)
    retry.py             Async retry with exponential backoff

Development

# Install dev dependencies
pip install -r requirements.txt

# Run linting
ruff check src/

# Run tests
pytest tests/

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选