Summary
Connects Claude to the atest API testing framework via gRPC, letting you run and manage API tests directly from conversations. You'll need atest running locally with its gRPC server exposed, then point this MCP server at that address. Supports both stdio and SSE transport modes depending on your setup. Useful when you're iterating on API test suites and want Claude to execute tests, check results, or help debug failing endpoints without switching contexts. The integration assumes you're already using atest for API testing and want conversational access to your test store and runner.
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →This is the MCP server of atest.
Get started
Please start the MCP server with atest gRPC port.
atest-store-mcp server --runner-address 127.0.0.1:64385
or start in Docker
docker run -p 7845:7845 ghcr.io/linuxsuren/atest-mcp-server --runner-address 127.0.0.1:64385
or start in npx
npx atest-mcp-server-launcher@latest server --mode=stdio --runner-address=localhost:64385
You can also set the MCP server mode with the --mode flag.
atest-store-mcp server --runner-address 127.0.0.1:64385 --mode=[sse|stdio]
MCP Server
{
"mcpServers": {
"atest": {
"name": "atest",
"type": "streamableHttp",
"description": "The MCP server of atest",
"isActive": true,
"baseUrl": "http://localhost:7845",
"disabledAutoApproveTools": []
}
}
}
or as stdio mode:
{
"mcpServers": {
"atest": {
"name": "atest-mcp-stdio",
"type": "stdio",
"description": "",
"isActive": true,
"command": "atest-store-mcp",
"args": [
"server",
"-m=stdio",
"--runner-address=localhost:64385"
]
}
}
}
How to build
You can build the binary with the following command:
make build
You can build .dxt package with the following command:
npm install -g @anthropic-ai/dxt
dxt pack
How to publish
You can publish the package with the following command:
npm login --registry https://registry.npmjs.com
npm publish --registry https://registry.npmjs.com
mcp-publisher login github
mcp-publisher publish
read more about mcp-publisher
Featured
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →Categories
Registryactive
Packageatest-mcp-server-launcher
TransportSTDIO
UpdatedSep 9, 2025
