Mutation Tester
Test the quality of your tests. Generate code mutations and find gaps where your test suite fails to catch bugs.
Get the CLI Tool
Run the mutation tester locally as an MCP server, or try it online below.
npx @clinetools/mutation-test- 8 mutation operators — arithmetic, comparison, logical, boolean, return, remove, boundary, string
- Shows exact line and column for every mutation
- Configurable operator selection and mutation cap
- Mutation type breakdown with per-operator counts
- Zero config — just run with npx
How to Use It
Three ways to generate mutations — pick the one that fits your workflow.
Try Online
Paste source code below and see all possible mutations — no install needed.
Use via CLI
Run as a local MCP server and connect any MCP-compatible client.
Add to Cline / Claude Code
Add to your MCP settings so your agent generates mutations automatically.
MCP Client Configuration (Cline)
{
"mcpServers": {
"mutation-test": {
"command": "npx",
"args": ["@clinetools/mutation-test"]
}
}
}Claude Code Configuration
# In your project's .mcp.json:
{
"mcpServers": {
"mutation-test": {
"command": "npx",
"args": ["@clinetools/mutation-test"]
}
}
}Example Prompt: Find Weak Tests
// Prompt to your AI agent:
"Generate mutations for my validation module and
tell me which ones my tests would miss"
// The agent calls:
generate_mutations({
code: fs.readFileSync("src/validate.ts", "utf-8"),
operators: ["comparison", "boundary", "boolean"],
maxMutations: 30
})Try It Online
Paste source code and see what mutations could be applied to test your tests.
Source Code
Paste a function to generate all possible mutations
Paste code and click Generate Mutations to see all possible mutants.
Mutations
Understanding Mutation Testing
Mutation testing measures how well your tests actually detect bugs.
What is Mutation Testing?
Mutation testing introduces small changes (mutants) to your code — flipping operators, changing return values, removing statements — and checks whether your tests catch the change. If a test fails, the mutant is "killed." If tests still pass, you have a gap.
Equivalent Mutants
Some mutations produce code that behaves identically to the original — these are "equivalent mutants." They cannot be killed by any test because the behavior hasn't actually changed. They are a known limitation and should be excluded from scoring.
Mutation Score
Mutation score = killed mutants / total mutants. A score of 80%+ is excellent. Unlike line coverage which only measures execution, mutation score measures whether tests actually verify correct behavior. It is a stronger metric for test quality.
When to Use
Use mutation testing on critical business logic, security-sensitive code, and complex algorithms. It is too expensive to run on an entire codebase regularly, so target the code where correctness matters most. Pair with coverage tools for full visibility.
Test Your Tests
Add the Mutation Tester to your agent's toolkit and find the gaps your test suite misses.
View Plans