CodingDocumented overviewLast checked Oct 2, 2026

mpai: Pricing, Features & Best Use Cases

Open-source terminal tool that lets a teammate join your existing Codex or Claude Code session, with their own name on each prompt.

Who should use mpai

mpai is worth trying for pairs of macOS developers who already use Codex or Claude Code and want a teammate to join a live session with their own name on each prompt. Based on the vendor's documentation, not on a hands-on test.

Best for: Small dev teams using Codex or Claude Code

Also a good fit for

  • Cofounders sharing AI coding sessions

Consider another tool if labeled public alpha (version 0.4.19 at time of check)

Or if

  • Site lists macOS only, with Tailscale required
  • Requires Node.js 20+ for guests and Homebrew or Node.js on the host

mpai pricing and usage limits

Prices checked Oct 2, 2026 on the official US pricing page. Prices and limits change often.

Free access and trial conditions

The site's structured data lists price $0 as open source.

Not applicable.

Paid plans: what changes as you upgrade

Open source web

Free source (opens in a new tab)

Month

For: Teams pairing on AI sessions

Limits

  • Public alpha

Includes

  • Session invites
  • Named prompts

Usage caps, credits and extra costs

Codex and/or Claude Code access and Tailscale required.

Not stated.

What is mpai, and how does it work?

mpai is an open-source command-line tool (public alpha) that lets a named teammate enter an existing Codex or Claude Code conversation from their own terminal, instead of screen sharing or pasting transcripts. The host chooses which session to share and creates an invite scoped to that session. The site describes a private setup over Tailscale with no public relay, installed via Homebrew or npm on the host. A guest needs Node.js 20+ and pastes a printed command. Prompts from the guest are attributed to them in the native transcript.

mpai homepage — screenshot of the coding AI tool, Oct 2, 2026
mpai's official site: how the product presents itself. godfaddaai.github.io, captured Oct 2, 2026

What you provide and what you get back

You provide

  • Existing Codex or Claude Code session
  • Invite command

You get back

  • Shared attributed prompts in the session transcript

Where you can use it

Platforms
macos, cli
Account
Not required

Best use cases for mpai

Scenarios are worked out by our editors from the documentation; the actual result depends on your input and plan.

Pairing with a teammate inside an in-progress AI coding session

Who
Pair developers
You provide
Live AI session
Expected result
Teammate joins and prompts
Plan needed
Free
Limitation
macOS and Tailscale only source (opens in a new tab)

Handing off a Claude Code or Codex session without re-explaining context

Who
Teams handing off work
You provide
Running session
Expected result
Handoff without re-explaining
Plan needed
Free

Reviewing a live AI session transcript from another machine

Who
Reviewers
You provide
Session transcript
Expected result
Live view elsewhere
Plan needed
Free
Limitation
Host must share the session source (opens in a new tab)

mpai features and their limits

mpai homepage — screenshot of the coding AI tool, Oct 2, 2026
Feature overview on the official site. godfaddaai.github.io, captured Oct 2, 2026

Join an existing Codex or Claude Code session from the terminal

Lets a teammate enter a Codex or Claude Code session from their own terminal. Avoids screen sharing.

Access: Host installLimit: Alpha source (opens in a new tab)

Per-session invites limited to one chosen session

Creates an invite for the one session the host picks. Limits exposure.

Access: HostLimit: One session per invite source (opens in a new tab)

Prompts attributed to the named teammate

Tags the guest's prompts with their name in the transcript. Shows who asked what.

Access: GuestLimit: Needs names set source (opens in a new tab)

Private connection over Tailscale, no public relay

Connects over Tailscale with no public relay. Keeps traffic private.

Access: Both sidesLimit: Requires Tailscale source (opens in a new tab)

mpai pros and cons

Reasons to choose mpai

Check before paying for mpai

How to use mpai for sharing an AI coding session with a teammate

Set up mpai

  1. Install mpai on the host with Homebrew or npm.
  2. Run mpai start and enter both names.
  3. Pick the session to share.

Example: sharing an AI coding session with a teammate

  1. Copy the printed guest command.
  2. Teammate pastes it with Node.js 20+.
  3. Teammate sends a prompt.

Review and export the result

  1. Verify the prompt shows the teammate's name.
  2. Confirm only the chosen session is visible.

mpai API

API access and separate billing

mpai has no public API in its official documentation. No API described.

Privacy, data use and commercial rights

Summarized from the vendor's current policies; this is not legal advice and a vendor's badge is not an independent audit.

How your inputs and outputs are handled

Traffic stays on a private Tailscale network with no public relay.

Best mpai alternatives for different needs

"Better suited" is based on documented features and pricing, not on a head-to-head test.

GitHub Copilot: better suited to

Cursor: better suited to

Devin Desktop (formerly Windsurf): better suited to

Alternatives
AlternativeWhen to choose itPricingLimitation
GitHub Copilot——Compare
Cursor——Compare
Devin Desktop (formerly Windsurf)——Compare

Explore mpai further

If you are looking at mpai

mpai user reviews

No user reviews yet.

Used mpai? Be the first to share your experience.

Write a review

Reviews are checked by a moderator before they appear. Your email is never published. Plain text only.

Your rating (required)

Show what you made or what went wrong. Images are published only if the review is approved; location data is removed.

    mpai FAQ

    What does mpai do?

    It lets a named teammate join an existing Codex or Claude Code session from their terminal. source (opens in a new tab)

    What does mpai require?

    macOS, Tailscale, Homebrew or Node.js 20+ on the host and Node.js 20+ for the guest. source (opens in a new tab)

    Is mpai open source?

    Yes, the site describes it as open source, with MIT listed in its structured data. source (opens in a new tab)

    How are guest prompts shown in mpai?

    Each guest prompt is attributed to the teammate's name in the session transcript. source (opens in a new tab)

    Sources and review methodology

    Official sources and verification dates

    Not verified: Linux or Windows support; Privacy policy; Roadmap beyond alpha.

    How we documented and maintain this page

    Every fact about mpai above comes from the 2 official pages listed here, last compared on Oct 2, 2026; we have not run mpai ourselves, so it has no editorial score. How we check and re-check pages.

    Article updated Oct 10, 2026; prices checked Oct 2, 2026. Outbound links to vendors are plain links; we do not earn commissions. How we review · Report an error