One conversation the whole team is in, with the assistant

Valencia Street packet

Brad Today at 7:30a

Read the Valencia Street packet and tell me what changed since the June draft.

Reading it now. It is 214 pages, and I still have the June draft you gave me.

Maya Today at 7:33a

Whatever it says about the parking requirement, that is the part we need.

Read the August packet

sfplanning.org, 214 pages

Compared it with the June draft

Jun 12 vs tonight

parking minimum: 1 per unit → none
height limit: 55 ft → 65 ft

Two changes so far, and the parking one is the bigger: the minimum is gone for the whole corridor. Pulling the exact language for Maya

Writing a reply…

Watch the assistant work, and steer the answer before the end

The steps appear as it takes them. Add a note mid-answer and it takes the note.

See how it works

Named threads, with unread counts and the last thing said

A thread per piece of work, named. See which ones moved since you last looked.

See how it works

Board prep

Jenny Today at 7:30a

Before I go: the board deck is Thursday. Can you go through the Q3 numbers overnight and flag anything that does not match what we told them in June?

Will do. I have the June deck and the Q3 export. I will post here when I am through.

Done, 2:14 AM. Three things do not match: churn is 4.1% against the 3.5% we said, the Acme renewal is annual not quarterly, and headcount is 41 not 38. I have written each one up in the memory under Board prep, with the source for each number.

The workspace keeps working with the laptops closed

Hand it a job at night. Open the thread in the morning and the work is there.

See how it works

Outside

Claude Code AI

ChatGPT AI

Gmail account

The web search

asks answers

gbrain.io

around the server

1 Holds the keys, the permissions and the limits
2 Writes the record of every call
3 Keeps the machine, OpenClaw and gbrain up to date
4 Restarts it, and opens a shell when you ask

runs tends

The workspace's server

The memory notes

The skills running

OpenClaw runs the work

A shell, on request ssh

One machine per workspace, nobody else on it. gbrain.io is the layer around it: nothing reaches the server except through it, and it is what keeps the server, OpenClaw and gbrain current without you touching them.

A private server per workspace, kept up to date by gbrain.io

The memory and skills run on a machine of the workspace's own, wrapped and kept current by gbrain.io.

See how it works

Sign in with Google, or an email link, and no passwords

The whole team gets in with the accounts they already have. There is nothing to reset.

See how it works

Members

4

Admins

2

Invited

1

Invite the team, and decide who administers

Members use every workspace in the organization. Admins invite, remove, and pay.

See how it works

Run the workspace on the model you prefer, from Anthropic or OpenAI

Pick the model a workspace thinks with, and change it later. Everything else stays.

See how it works

Valencia Street packet

One conversation the whole team is in, with the assistant

Valencia Street packet

A colleague can add to a question mid-thread

The assistant knows who said what

More in Workspace

Valencia Street packet

Watch the assistant work, and steer the answer before the end

Named threads, with unread counts and the last thing said

Board prep

The workspace keeps working with the laptops closed

A private server per workspace, kept up to date by gbrain.io

Come on in.

Sign in with Google, or an email link, and no passwords

Invite the team, and decide who administers

Run the workspace on the model you prefer, from Anthropic or OpenAI

Spin up a gbrain

One conversation the whole team is in, with the assistant

Valencia Street packet

What it does

A colleague can add to a question mid-thread

The assistant knows who said what

More in Workspace

Spin up a gbrain