Test Inventory
Lets AI coding agents query a project's test suite, skips and affected tests without
- Pricing
- Open source
- Platforms
- CLI, vscode, API
- Runs
- Locally
- Version
- 0.10.0
- Listing updated
- today
Similar MCP servers
About Test Inventory
Test Inventory is an open-source MCP server for e-commerce. It runs locally on your machine and is started by your AI client.
The README and site describe an MCP server built on the test-inventory scanner. Instead of executing a suite, it reads test source files for Playwright, Vitest, Jest, Mocha, Jasmine, Cypress, WebdriverIO, node:test and Bun and answers questions about what exists, which tests are skipped or marked fixme, and which tests a given change could affect.
Agents call tools such as test_summary, find_tests, get_tests, scan_diagnostics and diff_inventories. Results are paged, and filters cover states, tags, title text, changed files and .only leftovers. Each scan re-reads the files, so answers follow edits made during a session.
The server is aimed at developers who use MCP clients like Claude Code, Cursor or VS Code and want test-suite context without a full run. The docs say the server executes no project code, writes no files and opens no network connections, though npx may download packages during installation.
Install
Run in a terminal
claude mcp add test-inventory -- npx -y test-inventory-mcpAdd to claude_desktop_config.json
{
"mcpServers": {
"test-inventory": {
"command": "npx",
"args": [
"-y",
"test-inventory-mcp"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"test-inventory": {
"command": "npx",
"args": [
"-y",
"test-inventory-mcp"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"test-inventory": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"test-inventory-mcp"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"test-inventory": {
"command": "npx",
"args": [
"-y",
"test-inventory-mcp"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"test-inventory": {
"command": "npx",
"args": [
"-y",
"test-inventory-mcp"
]
}
}
}Using several servers? Build one config file with the MCP config generator.
Tools this server provides (5)
Taken from the project documentation; the running server may differ.
- test_summary
- Counts by state, tag and annotation, and the configs with their projects.
- find_tests
- Lists tests matching states, tags, grep, related files, projects, only or withoutReason, with chosen fields.
- get_tests
- Returns every key of up to 20 tests by id, with scan diagnostics.
- scan_diagnostics
- Lists problems the scan found, such as files that failed to parse.
- diff_inventories
- Lists changes between two saved complete JSON scans.
What it does
- Test discovery without execution
- Lists tests across supported runners by reading their source files, so the suite is never run.
- Skip and fixme reporting
- Finds tests in skip or fixme states along with their conditions and recorded reasons.
- Change impact lookup
- Given changed files, returns the tests whose dependencies include them.
- Scan diagnostics
- Reports files the scan could not parse or read, so incomplete inventories can be spotted.
- Saved inventory comparison
- Diffs two saved JSON scans to show added, removed, renamed or moved tests and state changes.
- Paged structured answers
- Returns results 50 at a time with cursors, and lets the agent choose which record fields to include.
Good for
- Ask an agent which tests cover a feature before editing it
- List skipped or fixme tests and their reasons during test triage
- Find which tests a git diff could affect so only those are run
- Check for a stray .only left in the test files before committing
- Compare test inventories between two revisions of a project
Pricing
Open source
MIT licensed; free to self-host via npx.
Prices change. Check huldoser.github.io before you buy.
Test Inventory FAQ
Does the server run my tests?
No. According to the README, the running server executes no project code, writes no files and opens no network connections.
Which test runners are supported?
The README lists Playwright, Vitest, Jest, Mocha, Jasmine, Cypress, WebdriverIO, node:test and Bun test files.
How is it installed?
Clients start it with npx, for example npx -y test-inventory-mcp@0.10 --framework playwright, and the README gives setup for Claude Code, Cursor and VS Code.
Does it require Node.js?
Yes, the README specifies Node.js 22.12 or newer.
Is Test Inventory free?
Yes. Test Inventory is open source, so you can use and self-host it for free.
How do I add Test Inventory to Claude Code?
Run this command in your terminal: claude mcp add test-inventory -- npx -y test-inventory-mcp
What are the best alternatives to Test Inventory?
The closest alternatives to Test Inventory are Testing, Testdino MCP, MCP Server, MCP Test Runner and Test.
Where this comes from
- Description: written from the project's README, the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.