MCPAgentRE

一个用于从TAPD平台拉取数据并提供质量分析报告的MCP Server项目

python

11

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "onecuriouslearner-mcpagentre": {
      "command": "uv",
      "args": [
        "--directory",
        "D:\\MiniProject\\MCPAgentRE",
        "run",
        "tapd_mcp_server.py"
      ],
      "env": {
        "DS_KEY": "YOUR_DEEPSEEK_KEY",
        "SF_KEY": "YOUR_SILICONFLOW_KEY"
      }
    }
  }
}

You run an MCP server that connects to TAPD to fetch requirements and defects data, then analyzes and summarizes it for AI clients. It provides a cohesive set of tools to acquire, preprocess, search, visualize, and generate insights from TAPD data, all accessible through a local MCP server you start and query from your AI client.

How to use

You will start the MCP server locally and connect your AI client to it. The server exposes a suite of tools for retrieving TAPD data, refining it, performing precise searches, analyzing trends, vectorizing data for semantic search, and generating overviews and summaries. Your AI client can invoke these tools to obtain up-to-date TAPD stories and bugs, run analyses, and receive structured results.

How to install

Prerequisites: you need Python 3.10.x and the UV tool to run MCP components.

Install Python 3.10.x from the official source and ensure it is added to your PATH.
Install the UV tool to manage MCP components.
Create a project directory and place the MCP server files in it. For example, use a path like D:\MCPAgentRE.
Install required dependencies via the project’s packaging configuration.

uv venv
uv sync

Additional sections

Configuration options include TAPD API credentials and optional LLM API keys for advanced features like smart overviews and description preprocessing. The server itself runs as a standard local process and can be started via the MCP runtime utility.

Security notes: keep sensitive API keys and workspace identifiers in secure environments and avoid exposing api.txt or keys in public repositories.

Troubleshooting tips: ensure the MCP server path is correct, verify that the start command is executed with the proper working directory, and check the logs for any runtime errors.

Available tools

get_tapd_data

Fetches TAPD story and bug data from the TAPD API and saves a local integration, returning a count.

get_tapd_stories

Fetches TAPD project stories with pagination, returns JSON directly without local storage.

get_tapd_bugs

Fetches TAPD project bugs with pagination, returns JSON directly without local storage.

preprocess_tapd_description

Cleans HTML in descriptions, extracts text/links/images, and optimizes expression via LLM API to reduce size while preserving meaning (requires LLM API keys).

preview_tapd_description_cleaning

Previews the cleaned description results including compression ratios and extracted information without modifying the original data.

precise_search_tapd_data

Performs exact field matching searches on TAPD data for stories and bugs with options for data_type and case sensitivity.

search_tapd_by_priority

Filters TAPD data by priority levels to quickly identify high-priority items.

get_tapd_data_statistics

Provides distribution and statistics for TAPD data across stories and bugs.

analyze_time_trends

Analyzes time-based trends for stories or bugs with configurable time fields and chart types.

vectorize_data

Converts TAPD data into vector form for semantic search and analysis.

get_vector_info

Returns a simplified view of the vector database status and statistics.

search_data

Performs semantic search over vectorized TAPD data using natural language queries.

generate_fake_tapd_data

Generates synthetic TAPD data for testing and demonstration purposes.

generate_tapd_overview

Generates a concise overview report of the TAPD project using an LLM.

analyze_word_frequency

Analyzes TAPD data to find frequent keywords and generates a word cloud.

example_tool

A placeholder tool demonstrating how MCP tools are registered and invoked.