MCP-JaCoCo
MCP-JaCoCo is a server tool that converts JaCoCo code coverage reports into formats optimized for Large Language Models (LLMs), making AI-driven analysis easier and more effective.
Why It Matters
As AI and LLMs play a bigger role in software development, traditional code coverage reports—like those from JaCoCo—need a makeover. While great for humans, their XML format isn’t ideal for AI tools to process or analyze. MCP-JaCoCo bridges this gap by transforming these reports into LLM-friendly formats, unlocking powerful benefits for development teams:
- Quick, meaningful summaries of code coverage
- Easy identification of untested or poorly tested code
- Smart suggestions for new test cases
- Streamlined AI-assisted test planning
- Automated documentation of coverage results With MCP-JaCoCo, teams can tap into AI’s full potential, boosting efficiency and insight in testing workflows.
What It Solves
- Complex Formats: Simplifies JaCoCo’s dense XML reports for AI use
- Scattered Data: Pulls coverage metrics into one accessible place
- Slow Analysis: Cuts down on time-consuming manual reviews
- Integration Hurdles: Makes raw data play nicely with AI tools
Key Features
- Smart Conversion: Transforms JaCoCo XML reports into LLM-friendly JSON format
- Flexible Coverage Types: Supports multiple coverage metrics (instruction, branch, line, etc.)
- Efficient Processing: Fast and lightweight report processing
- Structured Output: Well-organized JSON format for easy AI consumption
- Customizable Analysis: Filter coverage data by specific metrics of interest
Installation
Install MCP-JaCoCo using uv with this configuration:
{
"mcpServers": {
"mcp-jacoco-reporter-server": {
"command": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"mcp",
"run",
"/Users/crisschan/workspace/pyspace/mcp-jacoco-reporter/mcp-jacoco-reporter-server.py"
],
"env": {
"COVERED_TYPES": "nocovered, partiallycovered, fullcovered"
},
"alwaysAllow": [
"jacoco_reporter_server"
]
}
}
}
Tool
jacoco_reporter_server
- Reads JaCoCo XML report and returns coverage data in JSON format
- Input:
- jacoco_xmlreport_path: Path to JaCoCo xml report path(jacoco.xml)
- Return:
- String, formatted JSON data containing coverage metrics
Example output format:
[
{
"sourcefile": "PasswordUtil.java",
"package": "com/cicc/ut/util",
"lines": {
"nocovered": [],
"partiallycovered": []
},
"branch": {
"nocovered": [],
"partiallycovered": []
}
},
{
"sourcefile": "UserServiceImpl.java",
"package": "com/cicc/ut/service/impl",
"lines": {
"nocovered": [
33,
67,
69,
71,
72
],
"partiallycovered": []
},
"branch": {
"nocovered": [
67
],
"partiallycovered": [
32
]
}
},
{
"sourcefile": "Constants.java",
"package": "com/cicc/ut/constants",
"lines": {
"nocovered": [],
"partiallycovered": []
},
"branch": {
"nocovered": [],
"partiallycovered": []
}
},
{
"sourcefile": "AuthException.java",
"package": "com/cicc/ut/exceptions",
"lines": {
"nocovered": [],
"partiallycovered": []
},
"branch": {
"nocovered": [],
"partiallycovered": []
}
},
{
"sourcefile": "UserService.java",
"package": "com/cicc/ut/service",
"lines": {
"nocovered": [],
"partiallycovered": []
},
"branch": {
"nocovered": [],
"partiallycovered": []
}
}
]
Recommend MCP Servers 💡
mcp-typescribe
An MCP server implementation enabling LLMs to work with new APIs and frameworks
@rawr-ai/mcp-filesystem
A Bun-based MCP server for filesystem operations with granular permission controls and enhanced functionality.
@pyroprompts/mcp-stdio-to-streamable-http-adapter
An MCP server that acts as a STDIO to Streamable HTTP adapter, enabling existing MCP clients to interact with Streamable HTTP MCP servers.
@mcpfun/mcp-server-ccxt
High-performance CCXT MCP server for cryptocurrency exchange integration
website-downloader
MCP server to download entire websites
mcp-server-notify
Send system notification when Agent task is done.
