Flaky History Mcp

io.github.mrfentmenPublished by GitHub account mrfentmenv1.0.0Updated Nov 5, 2026

Local aggregate test-history stability analysis without test names or raw result contents. Tools...

VerifiedSTDIODesktop onlyDeveloper ToolsData & Analytics

Overview

AI-generated overview

A local MCP server that reads bounded test-history exports and reports aggregate pass, fail, skip and unknown counts as a coarse stability signal.

What you can do
  • Analyze local test history: read bounded JSON, JSONL, CSV and simple JUnit-like XML files and return aggregate pass, fail, skip and unknown counts.
  • Keep output coarse: results omit test names, file names, raw records, exact versions and values.
  • Stay within scope: only local paths under a configured root are read, with dependency, build, Git, cache and vendor directories skipped.
When to use it
Useful when you want an assistant to summarize how stable a test suite looks from exported history files, without exposing individual test names or raw results. It is a signal only, not a CI runner, retry engine or statistical proof of flakiness.
Requirements
Runs as a local Node.js process installed from the npm package and started via the built entry point. It reads local files only, bounded by the FLAKY_HISTORY_ROOT environment variable. No accounts, API keys or network access are declared.
Before you install
It reads files from local disk, so point FLAKY_HISTORY_ROOT at a narrow directory containing only the history exports you intend to share. Output is aggregate and deliberately lossy, so do not treat it as proof of flakiness or as a substitute for running tests.

Installation

In SourceWeft

  1. Open Flaky History Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

flaky-history-mcp

Flaky History is a local MCP tool for turning small test-history exports into coarse stability signals. It accepts bounded JSON, JSONL, CSV, and simple JUnit-like XML files, then reports aggregate pass, fail, skip, and unknown counts.

Tool

  • analyze_test_history: aggregate local history without returning test names, file names, raw records, exact versions, or values.

Safety

  • Local paths only, bounded by FLAKY_HISTORY_ROOT and realpath checks.
  • Reads only bounded test and result-looking JSON, JSONL, CSV, and XML files.
  • Skips dependencies, build output, Git data, caches, and vendor directories.
  • This is a stability signal, not a CI runner, test retry engine, or statistical proof of flakiness.

Run

bash
npm installnpm run buildnode dist/index.js

Source: README.md at commit 0699aa5

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.0LatestSep 22, 2026

Score

40% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.