get_failed_pipelines_by_versionLABSCO SUMMARY
Test failures read by version, not by run.
The question a release raises is whether a failure is new, and this cuts the results sheet along that line — by version, by pipeline over time, by failure reason — so the comparison takes one question instead of an afternoon of filtering.
What it is
A Python MCP server over a Google Sheet of MicroShift test results. It groups failed pipelines by MicroShift version, aggregates failures across versions, tracks one pipeline over time, searches failure reasons, and compares two versions.
What you get
- Failed testing pipelines grouped by MicroShift version
- Aggregate failure statistics across all versions
- Failure trends for a named pipeline over time
- A search across the recorded failure reasons
- Two MicroShift versions compared against each other
- Data read from the sheet on each request rather than from a cached copy
Requirements
A Google Cloud service account with the Sheets API enabled, its client email and private key in .env, and the spreadsheet shared with that service-account address as a viewer. The spreadsheet id is a constant near the top of server.py — edit it there to point at your own sheet.
Setup effort
Build from source — git clone https://github.com/kasturinarra/mcp-test-scenarios-server.git
USE CASES
01
Deciding whether a failure is specific to one version
Group the failed pipelines by MicroShift version and see whether the pattern predates the release you are chasing.
02
Seeing how one pipeline has behaved across releases
Read that pipeline's failure trend over time rather than judging it by today's run.
03
Finding everywhere a particular error shows up
Search the failure reasons across all recorded tests, to see whether the same cause is behind several pipelines.
04
Getting the shape of the failures before a release meeting
Pull the aggregate statistics and a version-to-version comparison, so the discussion starts from counts.
5 tools
Listed in project docs. Your client may show a different list.
get_failure_summaryget_pipeline_failure_trendssearch_failure_reasonsget_version_comparisonBEFORE YOU INSTALL
What to know first
Tools you get
5 tools named in the project's documentation.
Requires
- GOOGLE_CLIENT_EMAIL in your environment
- GOOGLE_PRIVATE_KEY in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
git clone https://github.com/kasturinarra/mcp-test-scenarios-server.git
cd mcp-test-scenarios-server
pip install -r requirements.txt
cp env.example .envEdit .env and set GOOGLE_CLIENT_EMAIL and GOOGLE_PRIVATE_KEY using the values from your service account JSON file.
Open server.py, find the SPREADSHEET_ID constant near the top of the file and set it to your spreadsheet ID.
MCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"microshift-test-analyzer": {
"command": "python",
"args": [
"<YOUR_PATH_TO_MCP_TEST_SCENARIOS_SERVER>/server.py"
],
"cwd": "<YOUR_PATH_TO_MCP_TEST_SCENARIOS_SERVER>"
}
}
}The configuration on this page does not set GOOGLE_CLIENT_EMAIL and GOOGLE_PRIVATE_KEY; the project's instructions put them in a .env file. Check which folder the server reads that file from before you rely on it.
