MCP-Allure
MCP-Allure is a MCP server that reads Allure reports and returns them in LLM-friendly formats.
Motivation
As AI and Large Language Models (LLMs) become increasingly integral to software development, there is a growing need to bridge the gap between traditional test reporting and AI-assisted analysis. Traditional Allure test report formats, while human-readable, aren't optimized for LLM consumption and processing.
MCP-Allure addresses this challenge by transforming Allure test reports into LLM-friendly formats. This transformation enables AI models to better understand, analyze, and provide insights about test results, making it easier to:
- Generate meaningful test summaries and insights
- Identify patterns in test failures
- Suggest potential fixes for failing tests
- Enable more effective AI-assisted debugging
- Facilitate automated test documentation generation
By optimizing test reports for LLM consumption, MCP-Allure helps development teams leverage the full potential of AI tools in their testing workflow, leading to more efficient and intelligent test analysis and maintenance.
Problems Solved
- Efficiency : Traditional test reporting formats are not optimized for AI consumption, leading to inefficiencies in test analysis and maintenance.
- Accuracy : AI models may struggle with interpreting and analyzing test reports that are not in a format optimized for AI consumption.
- Cost : Converting test reports to LLM-friendly formats can be time-consuming and expensive.
Key Features
- Conversion : Converts Allure test reports into LLM-friendly formats.
- Optimization : Optimizes test reports for AI consumption.
- Efficiency : Converts test reports efficiently.
- Cost : Converts test reports at a low cost.
- Accuracy : Converts test reports with high accuracy.
Installation
To install mcp-repo2llm using uv:
{
"mcpServers": {
"mcp-allure-server": {
"command": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"mcp",
"run",
"/Users/crisschan/workspace/pyspace/mcp-allure/mcp-allure-server.py"
]
}
}
}
Tool
get_allure_report
Reads Allure report and returns JSON data
Input:
- report_dir: Allure HTML report path
Return:
- String, formatted JSON data, like this:
{
"test-suites": [
{
"name": "test suite name",
"title": "suite title",
"description": "suite description",
"status": "passed",
"start": "timestamp",
"stop": "timestamp",
"test-cases": [
{
"name": "test case name",
"title": "case title",
"description": "case description",
"severity": "normal",
"status": "passed",
"start": "timestamp",
"stop": "timestamp",
"labels": [
],
"parameters": [
],
"steps": [
{
"name": "step name",
"title": "step title",
"status": "passed",
"start": "timestamp",
"stop": "timestamp",
"attachments": [
],
"steps": [
]
}
]
}
]
}
]
}
