Quickstart

1. Create an account

Nebyoulous is in early access. Invited emails are approved when you sign up. Other new accounts wait for an admin to approve them. Until then the web app shows a pending screen, and the API, CLI, and MCP server answer 403 PENDING_APPROVAL.

2. Connect your AI assistant

The hosted MCP server needs no install. Add one URL to your client and sign in in the browser when it asks. In Claude Code:

claude mcp add --transport http nebyoulous https://api.nebyoulous.space/mcp

Then run /mcp inside Claude Code and choose Authenticate. See Connect a client for Claude Desktop, Cursor, VS Code, and others. Try a prompt like “List my integration sets” or “Which of my rigs uses the ASI2600MC?“

3. Install the CLI

The CLI handles the parts that live on your disk: camera cards and uploads. It ships as a tarball on each cli-v* GitHub release. The repository is private for now, so use the GitHub CLI to download the latest one:

TAG=$(gh release list -R deweyjose/lumigraph --json tagName \
  --jq '[.[].tagName | select(startswith("cli-v"))][0]')
gh release download "$TAG" -R deweyjose/lumigraph -p 'nebyoulous-cli-*.tgz'
npm install -g ./nebyoulous-cli-*.tgz
nebyoulous --version

The CLI keeps its tokens in the OS keychain. On Linux that needs libsecret (sudo apt install libsecret-1-0 on Debian and Ubuntu).

nebyoulous login
nebyoulous whoami

login opens an approval page in your browser. Approve it from your signed-in web session and the CLI stores the tokens. On a remote shell, use nebyoulous login --no-browser and open the printed URL yourself.

5. Sync your first frames

Look at what’s on a card first. scan only reads:

nebyoulous devices
nebyoulous scan /Volumes/NIKON_Z6

Then create an integration set and upload the lights in one command. Add --dry-run to see the plan without uploading:

nebyoulous sync \
  --from ./M31 --create "Andromeda 2026-05-24" \
  --include "*.fit,*.fits" --frame-type LIGHT

Run the same command again after an interruption and it skips the files that already finished. See Sync frames for the details.

Stack in the cloud

Map the frames, start a stack, and download the master.

MCP tool reference

Everything your assistant can do in the workspace.

Get started

Workspace

1. Create an account

2. Connect your AI assistant

3. Install the CLI

5. Sync your first frames

Next

Stack in the cloud

MCP tool reference

​1. Create an account

​2. Connect your AI assistant

​3. Install the CLI

​4. Sign in from the CLI

​5. Sync your first frames

​Next

Stack in the cloud

MCP tool reference

1. Create an account

2. Connect your AI assistant

3. Install the CLI

4. Sign in from the CLI

5. Sync your first frames

Next