Ag Bash runs on your own machine — the client starts it, so there is no endpoint to ping. 64 installs a week from npm. Last commit 17 Sep 2026.
Sandboxed AI-native bash with 70 agentic tools: run_bash, code edit/diff, WASM Python/JS.
We read the source, 18 h ago · rules 3dff92dd89df
Things with no honest explanation: a promise that contradicts the code, code that runs at install time while hiding what it does, data leaving the machine.
"169.254.169.254", // AWS / GCP / Azure / OpenStack IMDS
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.
exec(commandLine: string, options?: ExecOptions): Promise<BashExecResult>;
const fullPath = path.join(testDir, filePath);
system(cmd);
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
Code changes quietly between releases, and nobody reads the diff of a dependency. We do, on every release — watch Ag Bash and you get told the day something new turns up.
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 ag-bash -- npx -y @ag-bash/mcp-server
{
"mcpServers": {
"ag-bash": {
"args": [
"-y",
"@ag-bash/mcp-server"
],
"command": "npx"
}
}
}
[mcp_servers.ag-bash]
command = "npx"
args = ["-y", "@ag-bash/mcp-server"]
{
"mcpServers": {
"ag-bash": {
"args": [
"-y",
"@ag-bash/mcp-server"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"ag-bash": {
"args": [
"-y",
"@ag-bash/mcp-server"
],
"command": "npx"
}
}
}
Run AI-generated code safely. gVisor-isolated Python/JS sandboxes for any MCP client. Pay per run.
Execute code in 8 languages (Python, JS, TS, Go, Java, C++, C, Bash) in gVisor sandboxes.
WasmAgent MCP server — WASM-sandboxed code execution with capability manifests and Tasks API.
Build Apps and run code in 30 languages — sandboxed, with persistent sessions for agent loops.
Deterministic code maps for AI agents: 7 tree-sitter tools for TS/JS/Py — map, find, function_body
Native Weavatrix MCP: 64 read-only tools for code, n8n, Dify, agents, Mermaid, and Web3.
Run Python code in a secure sandbox without local setup. Declare inline dependencies and execute s…
Run the Grok Build CLI as MCP tools: agent runs with live progress, code review, web research.
Answers built from our own checks of this server.