On Glama.
Hands on your machines.
AI Commander is listed on Glama as a hosted Streamable HTTP MCP server — the remote-machine harness for Claude Code, Codex, ChatGPT, Cursor, and Grok. It is not an alternative to those tools.
Live Glama links: server listing · connector + TDQS · awesome-mcp-servers (Command Line).
Connect (same endpoint Glama indexes)
Endpoint: https://aicommander.dev/mcp (Streamable HTTP).
First connect may open a one-time browser sign-in (OAuth Dynamic Client Registration).
For a code-only first hour without that flow, use https://aicommander.dev/mcp?anonymous=1.
claude mcp add --transport http aicommander https://aicommander.dev/mcp
Client-specific harness pages (WITH, not alt): Claude Code · Codex · ChatGPT · Cursor · Grok. Full matrix: Connect your AI client.
What Glama visitors get
- Remote shell on Linux, macOS, and Windows — by session code or saved alias.
- Detached jobs that outlive the chat (builds, training, headless agents).
- Fleet + GPU facts so the agent picks the right box before it burns an hour.
- File courier (Pro) for checkpoints and diffs up to 100 MiB.
- Scored tool definitions on the Glama connector page (TDQS), MIT public plugin repo.
Not a replacement — a multiplier
| You already use | AI Commander adds |
|---|---|
| Glama / any MCP host | https://aicommander.dev/mcp → remote exec + jobs |
| Claude Code | MCP remote exec + jobs on your fleet |
| Codex / ChatGPT / Cursor / Grok | Same harness; see sibling /with-*/ pages |
FAQ
https://aicommander.dev/mcp.https://aicommander.dev/mcp. First connect opens a one-time browser sign-in (DCR). For a code-only first hour, use ?anonymous=1. See /howto/ for the full matrix.Next steps
Open the Glama listing, install the agent on a machine you control, connect your client, then ask for a real task on a remote box.