Skip to content
ychampionPublic

About

Open-source personal agent with its own computer. It remembers your work and keeps going while you're away. Works with any model.

Topics

Resources

Contributing

Security policy

Stars

18 stars

Watchers

0 watching

Forks

Latest commit

 

History

2,685 Commits

Folders and files

Repository files navigation

Melete

Personal superintelligence that runs as you.

Melete lives on your own computers and works in every account you're signed into: work and personal mail, both calendars, your files, your browser, your servers. It keeps watch over all of it, day and night, notices what needs you, and takes care of it.

Melete is in early beta. You can run it yourself today.

Learn more at melete.si. Hosted Melete: join the waitlist at waitlist.melete.si.

Try it

With Docker running, one command installs and starts Melete:

curl -fsSL https://raw.githubusercontent.com/ychampion/melete/main/install.sh | bash

On Windows with Docker Desktop, in PowerShell:

irm https://raw.githubusercontent.com/ychampion/melete/main/install.ps1 | iex

Then open http://localhost:3101 and create your account. It starts with a practice model, so you can look around right away; add your own key in Settings → Models. Run the command again to update. Run it yourself shows the longer manual path, and Deployment lists the installer's options.

A day with Melete

2:10 a.m. It notices. The card on your domain renewal was declined, and the notice went to an old personal address you rarely open. Melete watches every inbox you connect, so by the time you're up the renewal page is open in your browser, waiting for you to confirm the new card.

8:30 a.m. Two calendars, one life. The dentist moved your appointment, and it now lands on a board call in your work calendar. Melete sees the overlap across both calendars and has a reschedule drafted from the right address.

11:00 a.m. Things only you are signed into. Your accountant asks for last quarter's numbers. Melete opens your bank in your own browser, where you're already signed in, reads the statements and lines them up against the invoices in your Drive. Two-factor prompts and logins work the way they do for you, because it's your browser.

4:45 p.m. Say it once. "Dinner for six on the 9th, somewhere the kids will like." Melete keeps hold of it until it's booked, and tells you plainly if time is running short.

On the drive home. You talk, it answers out loud, and it keeps working while you speak.

Why it runs as you

  • One agent across your whole world. Google, Microsoft, iCloud and any IMAP mailbox, your calendars and Drive, any app with an MCP server, and your own computer, all in one place, for one agent that sees how they connect.
  • Your sign-ins, your devices. The companion on your computer lets Melete use your browser signed in as you, the folders you share and the commands you approve, and it does only what both you and that computer allow.
  • It can afford to watch everything. It runs on hardware you already have, with the model you choose, including one on your own machine. When nothing changes, nothing runs and nothing is spent.
  • Its own computer too. Code, commands and long jobs run in a sandbox on E2B, Modal or Daytona, and a separate browser handles web forms, so heavy work never ties up yours.
  • You can check its work. It asks before anything leaves your hands: a message to someone, a payment, a delete. Everything else leaves a receipt, with Undo while it still works. You can see, correct or forget anything it remembers, and a privacy router keeps account numbers and private conversations away from cloud models.

Melete following a job through: the draft, the one approval, then the replies and follow-up until it's settled.

Run it yourself

You need Docker, Bun and an API key for your model provider.

git clone https://github.com/ychampion/melete.git && cd melete && bun install --frozen-lockfile
read -rs FIREWORKS_API_KEY && export FIREWORKS_API_KEY; bun run deploy/scripts/configure.ts; unset FIREWORKS_API_KEY
docker compose -f deploy/docker-compose.yml up -d --build --wait --wait-timeout 300

The second line waits for you to paste your Fireworks API key and press Enter. The key stays hidden, configure.ts writes it into deploy/.env, and the line then clears it from your shell. For another provider, export its key instead and name the provider and model, for example bun run deploy/scripts/configure.ts --provider anthropic --model <model id> with ANTHROPIC_API_KEY. Then open http://localhost:3101 and create your account.

Once your account exists, give the agent a browser of its own to read pages and fill in forms with:

bun run melete browser enable

It sets up the browser worker for your space and starts it. From then on, add -f deploy/docker-compose.browser.yml after -f deploy/docker-compose.yml on a docker compose command.

To try a demo with a practice model first, use bun run deploy/scripts/configure.ts --fake as the second line. It needs no key, and you can switch to your own model later.

On a server, you can skip the build and pull the published images instead: Using prebuilt images.

bun run melete check, bun run melete doctor and bun run melete status judge an installation and name what to do next: The melete command.

Deployment covers version requirements, Windows, remote servers, HTTPS, Tailscale, backups and removal.

Set it up with your coding agent

Claude Code, Codex, Cursor or another coding agent can install Melete for you, on this computer or on a server. It checks the machine, starts Melete, and tells you when to create your account. You type your own passwords and keys, into Melete or your own terminal, and the agent does the rest. Paste this:

Set up Melete on this machine using https://github.com/ychampion/melete/blob/main/SETUP-WITH-AN-AGENT.md

The guide also covers the optional parts: a computer for the agent, voice, mail and calendar, and a public address for other assistants. For Claude Code there is a skill you can install once.

Connect your model

Pass the provider to configure.ts with --provider and --model, and export the key it reads first:

Provider Key it reads
fireworks (the default) FIREWORKS_API_KEY
anthropic ANTHROPIC_API_KEY
openai OPENAI_API_KEY
google GOOGLE_API_KEY
openai-compatible OPENAI_COMPAT_BASE_URL and OPENAI_COMPAT_API_KEY, which can point at a model server on your own network
chatgpt No key. Once Melete is running, you sign in to your ChatGPT account under Settings → Models, or through its sign-in routes.

You can also connect a provider from the app, in Settings › Models: paste a key, test it, and choose the model, with no restart. Write the model's name the way the provider does. To change provider or model later, edit MELETE_DEFAULT_PROVIDER, MELETE_DEFAULT_MODEL and the key in deploy/.env, then restart the two services that use them:

docker compose -f deploy/docker-compose.yml up -d --force-recreate --wait melete runtime

Providers explains each one, including signing in with ChatGPT.

Connections and plugins

  • Mail and calendars. Add them in Settings → Connections. Gmail and iCloud take an app password, and other IMAP and CalDAV accounts take their password. With your own Google or Microsoft app set in deploy/.env, Gmail, Google Calendar and Outlook connect by signing in (Google setup, Microsoft setup).
  • Any MCP server. Add one over HTTP, or from a package or image, and a packaged server runs in its own locked-down container. A starter set of files, fetch, time and GitHub is listed at GET /plugins, and each installs with one request.
  • A sandbox. Connect an E2B, Modal or Daytona account as a Sandbox connection, and the agent's terminal runs there, with a receipt for each command.
  • A browser. The browser worker fills forms in its own isolated browser.
  • Your own computer. Pair it in Settings → Devices to share folders, approve commands, and let Melete use your browser signed in as you (devices).

Security

  • Every message it sends waits for your approval, or falls under a limit you set for someone you trust. Approving a chase's first message also covers up to three follow-ups that repeat it to the same person, and you can withdraw that in Settings → Rules.
  • The passwords and sign-in tokens it stores are sealed with your installation's master key.
  • The agent's code runs in an isolated container that can reach only Melete's own service, which checks each action against what you allowed.
  • Each person on an installation has their own space, and their jobs, drafts and receipts stay private to them.
  • Every action leaves a receipt, and one that can be reversed shows an Undo button while it still works.
  • To report a vulnerability, follow SECURITY.md. The threat model has the details.

More it can do

  • Talk to it: push-to-talk and a hands-free voice mode, with one ElevenLabs key (voice).
  • Give it long work, and it keeps going in shifts for as long as it takes (long work).
  • Use it from ChatGPT, Claude, Claude Code, Hermes or OpenClaw as a connector (Melete as an MCP server).
  • Teach it by correcting it, then see, pause or remove what it learned in Settings → Memory.
  • Start your day with a short brief of your calendar, your tasks and the decisions waiting on you.
  • Correct or forget anything it remembers, and a forgotten fact stays gone after a restore.

Documentation

Licence and credits

Melete is Apache-2.0. See LICENSE and NOTICE.

Melete's agent runtime is built on Hermes Agent by Nous Research, customised for Melete.

About

Open-source personal agent with its own computer. It remembers your work and keeps going while you're away. Works with any model.

Topics

Resources

Contributing

Security policy

Stars

18 stars

Watchers

0 watching

Forks

Releases

Contributors

Languages