Keep Junie.
Give it your machines.
AI Commander is not an alternative to JetBrains Junie, Claude Code, Codex, ChatGPT, Cursor, GitHub Copilot, Sourcegraph Cody, Grok, Gemini, Windsurf, Continue, Cline, Roo Code, Zed, OpenCode, Qwen Code, goose, Amp, Kilo Code, OpenHands, Amazon Q Developer, Augment Code, or Devin. It is the remote-machine harness those tools plug into over MCP: outbound-only shell, detached jobs that outlive the chat, and file transfer on hardware you already own.
- Install the host agent on the machine you want your AI to control. It shows a private machine code.
- Connect your AI: paste this prompt, then replace
AIC-…with your code —Help me connect this AI client to AI Commander using https://aicommander.dev/start.md. Then check the status of my machine: AIC-…— or add the connector URLhttps://aicommander.dev/mcpand sign in with an email code (per-client steps).
Sibling harness pages: Cursor · GitHub Copilot · Cody · OpenHands · Amazon Q · Augment · Devin · Aider · Cline · Roo Code · Zed · OpenCode · Qwen Code · goose · Amp · Kilo Code · Continue · Windsurf · Claude Code · Grok · Codex · ChatGPT · Gemini · Cua Spaces · Kiro · Warp.
Connect Junie (remote MCP)
Junie, the coding agent by JetBrains
(Junie CLI MCP docs
·
IDE MCP Settings),
reads MCP servers from an mcp.json file. Per JetBrains’ docs, Junie CLI uses the same MCP JSON configuration as Junie in JetBrains IDEs,
and remote servers are declared with a url (plus optional headers).
Point it at AI Commander’s hosted MCP endpoint:
{
"mcpServers": {
"aicommander": {
"url": "https://aicommander.dev/mcp"
}
}
}
Use ~/.junie/mcp/mcp.json for user scope (all projects) or .junie/mcp/mcp.json at the project root for project scope.
Do not commit secrets: if you add an Authorization header with an API key, keep it in the user-scope file.
In Junie CLI, the /mcp screen lists servers and their status; remote servers that need OAuth show
Authorization required until you run Authorize and finish the browser sign-in.
In a JetBrains IDE, open MCP Settings for Junie and edit the same JSON.
Prefer a local stdio bridge?
{
"mcpServers": {
"aicommander": {
"command": "npx",
"args": ["-y", "@aicommander/mcp"],
"env": { "AICOMMANDER_SERVER": "https://aicommander.dev" }
}
}
}
Full client matrix (Junie, Augment, Devin, Claude Code, Codex, ChatGPT, Cursor, Copilot, Cody, Grok, Gemini, OpenCode, Qwen Code, goose, Amp, Kilo Code, Continue, Windsurf, Cline, Roo Code, Zed, OpenHands, Amazon Q, REST): Connect your AI client.
First connect may open a one-time browser sign-in (OAuth Dynamic Client Registration).
For a code-only first hour without that flow, append ?anonymous=1 to the MCP URL.
For optional account features (saved aliases), pass a Bearer API key in headers.
Menu names and options can differ between Junie versions — check the linked JetBrains docs for your build.
Long commands: Junie CLI’s 60-second MCP timeout
Per JetBrains’ Junie CLI MCP docs (checked 2026-10-25),
Junie CLI waits up to 60 seconds for an MCP server to respond to each request, including every tool call. A build or test run on a remote box can take longer than that,
so for a remote_exec call that takes a few minutes you can raise the optional timeout field on the server. It is in milliseconds, between 1000 and 3600000;
values outside that range are ignored and the 60-second default applies.
{
"mcpServers": {
"aicommander": {
"url": "https://aicommander.dev/mcp",
"timeout": 300000
}
}
}
For work that should outlive any single request — full test suites, training, large builds — start a detached job with remote_job_start and poll it with remote_job_status,
so the run does not depend on Junie keeping one request open. If you use project scope, note that interactive Junie CLI asks you to trust a project before it loads project-scoped
inputs such as MCP servers (Junie docs).
What you get that the chat alone cannot
- Remote shell on Linux, macOS, and Windows — by session code or saved alias.
- Detached jobs for builds, test suites, and training on the box that has the repo.
- Fleet list + GPU facts so Junie picks the right machine before it burns an hour.
- File courier (Pro) for checkpoints and diffs up to 100 MiB.
Junie Remote mode and AI Commander
Junie CLI (in Early Access) has a Remote mode: run /remote, choose Start Remote Session, and open the printed junie.jetbrains.com link to see the running session in a browser, so you can keep working from another device while the process stays on your machine — which must stay awake (Junie docs).
For unattended work, Junie CLI also runs headless in CI pipelines (Junie docs) and through the Junie GitHub Action on your GitHub runners (Junie docs).
AI Commander works alongside both: Junie still decides what to do, and AI Commander gives it MCP tools to reach your other machines — servers, GPU rigs, NAS, Windows boxes — through a host agent that only connects outbound, with no SSH keys or open ports, plus detached jobs that keep running after the chat ends.
Not a replacement — a multiplier
| You already use | AI Commander adds |
|---|---|
| JetBrains Junie | mcpServers.aicommander.url → https://aicommander.dev/mcp in ~/.junie/mcp/mcp.json |
| Augment Code | auggie mcp add aicommander --transport http --url https://aicommander.dev/mcp |
| Devin | Settings → MCP → custom HTTP → https://aicommander.dev/mcp |
| Amazon Q Developer | type: http → https://aicommander.dev/mcp (or npx -y @aicommander/mcp) |
| Claude Code | claude mcp add --transport http … → remote exec + jobs |
| Codex / ChatGPT | Same MCP endpoint or REST /api/v1/exec |
| Cursor / Continue / Windsurf | npx @aicommander/mcp stdio bridge |
| Copilot / Cody | HTTP MCP / cody.mcpServers + bridge |
FAQ
mcp.json: user scope ~/.junie/mcp/mcp.json or project scope .junie/mcp/mcp.json, with mcpServers.aicommander.url set to https://aicommander.dev/mcp. In Junie CLI you can also use the /mcp screen; in JetBrains IDEs open MCP Settings for Junie. Prefer stdio? Use command npx with args -y @aicommander/mcp. First connect may need a one-time browser sign-in (OAuth). Use ?anonymous=1 for a code-only first hour. See /howto/ for the full matrix.remote_exec. Set the optional timeout field on the aicommander server in mcp.json (milliseconds, 1000–3600000) to allow remote_exec calls that take a few minutes, or start a detached job for anything longer and check on it afterwards. Source: Junie CLI MCP docs, checked 2026-10-25.Next steps
Install the agent on a machine you control, add AI Commander to Junie’s mcp.json (remote URL or npx -y @aicommander/mcp), then ask Junie for a real task on a remote box.