CARBON runs on your own machine — the client starts it, so there is no endpoint to ping. 211 installs a week from npm. Last commit 11 Oct 2026.
AI testing harness with Jay: risk-based checks, bug investigation, and local evidence reports.
We read the source, 1 d ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
этот файл ставится пользователю, но в репозитории его нет
proc = subprocess.run(
…m</option><option value="large">Large (+$0.75)</option></select></div><div class="field"><label for="milk">Milk</label><select id="milk"><option value="regular">Regular</option><option value="oat">Oat (+$0.75)</option><option value="almond">Almond (+$0.75)</option></select></div>…
const dir=path.join(home,'.config/carbon/usage',product.edition), config=path.join(home,'.config/carbon/config.json');
else{go('home');return}$('#main').innerHTML=html;bind(page)}
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add carbon -- npx -y @testers-ai/carbon-mcp
{
"mcpServers": {
"carbon": {
"command": "npx",
"args": [
"-y",
"@testers-ai/carbon-mcp"
]
}
}
}
[mcp_servers.carbon]
command = "npx"
args = ["-y", "@testers-ai/carbon-mcp"]
{
"mcpServers": {
"carbon": {
"command": "npx",
"args": [
"-y",
"@testers-ai/carbon-mcp"
]
}
}
}
{
"servers": {
"carbon": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@testers-ai/carbon-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
CARBON_ANALYTICS (Optional content-free analytics. Off by default; set off to force-disable), DO_NOT_TRACK (Set 1 to disable analytics regardless of preferences).
Local QA workflows, test evidence, screenshot maps and confidence reports for AI coding agents.
Report local task-scoped test and build residue without cleaning resources.
Deep Research MCP for multi-round investigations, traceable evidence, and cited reports.
FMCSA motor-carrier risk scores, monitoring, and carrier-selection evidence reports for AI agents.
Configure mobile bug reporting and investigate retained reports with your coding assistant.
PDF, photo, email, and file comparison evidence checks with plain-language reports.
Accessibility scans with evidence, public reports, and fresh checks after fixes.
Free reproducible Python integration fixes with execution evidence and result reporting.
Answers built from our own checks of this server.