Pitest MCP server: nobody has built one

What the only community project actually does, why mutation testing is an awkward fit for a connector, and what to use instead.

Last verified 25 August 2026 · from pitest.org, the hcoles GitHub organization, the MCP registry and the GitHub API

This page is one of 90 in a directory of monitoring and developer tools. Each page checks what an AI coding agent can get out of the tool through its MCP server, the connector that lets the agent query the tool directly.

What is the Pitest (PIT) MCP server?

PIT is a mutation testing tool for Java. It changes your code in small ways, one mutation at a time, then runs your tests to see whether they notice. A mutation nothing catches is a gap in the test suite.

The output is a mutation score and a report listing which mutants survived, in which class and on which line. That report is HTML, written to disk by the build.

A connector would let an agent ask which mutants survived last night. What exists instead parses the HTML afterwards, which is the same information a build artifact already carries.

Everywhere a PIT connector would be, and what is there:

The last row is a report reader, not a PIT connector. It needs a completed run and a file on disk, and knows nothing about PIT itself.

Mutation testing sits in code, tests and review, and reaches it only through a file.

PIT exposes a mutation score only after a build produced one and a parser is installed. There is no route from a question to a fresh run.

What can an AI agent do with Pitest (PIT)?

PIT has no connector, so nothing calls it. The only project found reads the HTML report that PIT has already written, which means an agent can ask about a mutation run that finished and cannot start one. It publishes no tool list, so its tool names are whatever analyzer.py registers when you run it. The mutation run itself stays in the build, where it was.

Pitest (PIT) MCP server limits: where answers come back incomplete

There is no PIT connector. The first limit below explains what the one project actually does.

How to configure Pitest (PIT) MCP for agents

1. 1

   Decide whether you need a connector at all

   The build already writes the report, and agents can read files.

2. 2

   Do not expect to trigger a run

   Nothing here starts PIT. That stays a build step, invoked by Maven, Gradle or your pipeline.

3. 3

   Weigh the dependency

   The one project has no stars, and adding it puts an unreviewed parser in the loop.

4. 4

   Point the agent at the report instead

   target/pit-reports is a path, and reading it needs no server.

Paste this into your agent’s instructions

A mutation score without its run is a number with no meaning. The block ties every figure to the report that produced it.

When you answer about mutation testing, state:
- The report file you read and when the run that produced it
  finished. A mutation score describes one run.
- Which classes the run covered. PIT is usually scoped to a
  package, not the whole codebase.
- That nothing here can start a run, so the newest figure
  available is the last build's.
Never state that data does not exist. State the report you
read and what it contained.
Write your answer in ASD-STE100 Simplified Technical
English. Use short sentences with one idea in each.

Do you need the MCP server at all?

No. The report is a file, and an agent with file access can read target/pit-reports without any server.

PIT writes XML and CSV alongside the HTML, and both parse more reliably than scraping a rendered page.

Stryker Mutator covers the same idea for JavaScript, TypeScript, C# and Scala, and has a connector story of its own.

For test quality more broadly, SonarQube and BuildPulse both publish servers and answer coverage and flakiness questions directly.

Read the report file. A server that parses HTML you already have adds a dependency and no reach.

Pitest (PIT) MCP server setup

There is no PIT server. The only option found is a third-party report reader, and what it reads is HTML that PIT has already produced.

Set expectations before configuring anything. It cannot run mutation testing and it cannot reach PIT itself, so the mutation run stays exactly where it is, in your build.

It is also undocumented. The repository carries no README at all, only Python files and a license, so the steps below are inferred from the repository contents rather than from instructions.

mvn org.pitest:pitest-maven:mutationCoverage
# writes target/pit-reports/

{"mcpServers":{"pit-report":{
  "command":"python",
  "args":["<path>/analyzer.py"]}}}

Every other client

Each block below is the configuration for one client, with the file path and the key that client expects.

gh api repos/Barry2llen/TestReportHtmlAnalyzerMCP/contents \
  --jq '.[].name'    # no README

Read-only access, permissions and security

• The one project has no stars and no reviewers

  Adding it means running unreviewed code that parses build output.

• Reports describe your source

  A mutation report names classes, methods and line numbers, so it is a map of the codebase.

• Nothing writes

  A report parser reads files. It cannot change code, tests or a build.

• PIT itself modifies bytecode

  That happens inside the build, under your own control, and no connector changes it.

Troubleshooting

You cannot find a PIT MCP server

There is not one. pitest.org documents none, the hcoles organization has no MCP repository, and the registry returns zero results.

The analyzer returns nothing

It needs an HTML report that already exists. Run PIT first, then point it at the output directory.

You want the agent to run mutation testing

Nothing here can. Invoke PIT through Maven or Gradle in the build, then read the report it produces.

Arcmutate features are missing

Arcmutate is the commercial extension set for PIT, and its sitemap contains no MCP references either.

Pitest (PIT) MCP server: Reference

What engineers report

There is nothing to report about a PIT connector, because none exists. The counts below are the searches that establish it.

“a stdio mcp server that analyzes html created by jacoco and pitest”

On 25 August 2026 the MCP registry returned zero results for pitest, a GitHub search of the hcoles organization for MCP repositories returned zero, and the Arcmutate sitemap contained no MCP references. pitest.org documents no MCP server. The single community project's facts come from the GitHub API on the same date.

Should you connect an agent to Pitest (PIT)?

No. Only community servers exist, and none of them is usable. Best for reading target/pit-reports directly, with mutation testing wired into CI.

Strongest fit

Nobody, at present. Teams running PIT are better served reading the report the build already writes.

Main advantage

The one project is MIT licensed and small enough to read before trusting.

Main weakness

It parses output rather than reaching PIT, so it adds a dependency without adding reach.

Operational risk

Low. It reads files and writes nothing, and the risk is depending on an unreviewed project.

Frequently asked questions