get_allure_reportLABSCO SUMMARY
Turn last night's Allure run into something an assistant can reason about.
The report is already on disk when CI finishes; this reads it back as structured suites and statuses, so failure patterns can be grouped, summarised and explained without anyone scrolling through the generated HTML.
What it is
A single-tool server that reads an Allure HTML report off disk and returns it as formatted JSON: test suites with their names, titles, descriptions, status and start and stop timestamps, in a shape a model can work through.
What you get
- An Allure report read from the directory path you pass as report_dir (get_allure_report)
- The run returned as JSON: suites, titles, descriptions, pass and fail status, start and stop timestamps
- A local stdio server run under uv, reading reports already on the machine
Requirements
uv and the mcp[cli] package, plus the path to a generated Allure report on the same machine. It reads a report directory; it does not run your tests.
Setup effort
Clone and run — clone the repository, then add the configuration for your client; uv runs the script from your clone, with nothing to build
USE CASES
01
Summarising a test run someone else has to read
Point it at the report directory and the whole run comes back as suites with statuses and timings, which is enough to write the paragraph that goes into the standup or the release ticket.
02
Finding the pattern behind a batch of failures
With every suite, title and status in one structured payload, the assistant can look for what the failures share, rather than you opening twelve report pages to notice they are all the same fixture.
03
Comparing a cycle against the last one
Read two report directories in turn and compare the suite-level results and durations, so a regression or a slowdown shows up as a difference instead of a feeling.
1 tool
Listed in project docs. Your client may show a different list.
BEFORE YOU INSTALL
What to know first
Tools you get
1 tool named in the project's documentation.
Requires
- uv on PATH
- Python 3.11+
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Clone the repository, then add the configuration for your client; uv runs the script from your clone, with nothing to build.
Server setup
TERMINAL
git clone https://github.com/crisschan/mcp-allure.gitMCP Clients
Any MCP client
ANY MCP CLIENT · JSON
{
"mcpServers": {
"mcp-allure-server": {
"command": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"mcp",
"run",
"<YOUR_PATH_TO_MCP_ALLURE>/mcp-allure-server.py"
]
}
}
}<YOUR_PATH_TO_MCP_ALLURE> is the full path of your clone, so the last argument points at mcp-allure-server.py inside it. uv run --with mcp[cli] pulls in the MCP SDK when your client starts the server, so there is nothing else to install.
