MCP MCP Server

MCP server for Gaffer test history - give your AI assistant memory of your tests. The mcp mcp server wraps that behind the Model Context Protocol, so an assistant can use it through 14 defined tools rather than through you.

What it actually does

This MCP server connects AI coding assistants like Claude Code and Cursor to your Gaffer test history and coverage data. It allows AI to:

Its toolset

Everything the assistant can do here goes through one of these:

• list_projects — List all projects you have access to
• get_project_health — Get health metrics (pass rate, flaky count, trends)
• list_test_runs — List recent test runs with optional filtering
• get_test_run_details — Get parsed test results for a specific test run
• get_report — Get report file URLs for a test run
• get_report_browser_url — Get a browser-navigable URL for viewing reports
• get_test_history — Get pass/fail history for a specific test
• get_flaky_tests — Get tests with high flip rates (pass↔fail)
• get_slowest_tests — Get slowest tests by P95 duration
• compare_test_metrics — Compare test performance between commits
• get_failure_clusters — Group failed tests by root cause (error similarity)
• get_upload_status — Check if CI results are uploaded and processed

Configuration

You will need 2 environment variables: GAFFER_API_KEY, GAFFER_API_URL. The server will not start without them, which is usually why the tools fail to appear on a first run. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.

1. A Gaffer account with test results uploaded 2. An API Key from Account Settings > API Keys

Adding it to your client

The server ships on npm as @gaffer-sh/mcp, so your MCP client can launch it on demand — there is no separate build step. Add the server block to your client's configuration, restart it, and the tools register themselves.

When to reach for it

Among the developer tooling options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. MCP's toolset — list_projects, get_project_health, list_test_runs and 11 more — is a fair guide to whether it matches your workflow. It is maintained by gaffer-sh; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against MCP's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.

Caveats

• It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
• With 14 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch MCP.
• Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
• MCP clients confirm each tool call by default. Leave that on until you have watched what the mcp mcp server does with a few real requests.