Tewip runs on your own machine — the client starts it, so there is no endpoint to ping. 112 installs a week from npm. Last commit 10 Oct 2026.
Proves software work: triages failing tests, checks whether a test can fail, and reads migrations.
Today is the operative word: we check Tewip every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
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 tewip -- npx -y tewip-ai
{
"mcpServers": {
"tewip": {
"args": [
"-y",
"tewip-ai"
],
"command": "npx"
}
}
}
[mcp_servers.tewip]
command = "npx"
args = ["-y", "tewip-ai"]
{
"mcpServers": {
"tewip": {
"args": [
"-y",
"tewip-ai"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"tewip": {
"args": [
"-y",
"tewip-ai"
],
"command": "npx"
}
}
}
This one needs environment variables set before it will start:
OPENAI_API_KEY (Optional. Turns on the AI layer for failures the deterministic rules cannot decide. Without it only the rules run, and Tewip says which layer decided each answer.).
The author declared them in the registry entry; get the values from the project itself.
Check whether a Claude Code skill was tested and whether it works before installing it.
AI software engineer — writes code, opens PRs, reviews code, generates tests, and more.
Ask your AI agent why your tests are failing. Instant analysis on failures and regressions.
MCP Server for software testing, test case management, test execution, and defect tracking.
Human workforce for AI agents: field checks, user testing and device tests. Operator-reviewed work.
Make a coding agent prove it: run the repo's real tests, typecheck and lint before it says done.
Self-healing CI tests and proof of work for AI agents
Triage failing DeviceCloud Maestro test runs: fail reasons, screenshots, flaky-flow analytics
Answers built from our own checks of this server.