Padar

Claude Code · Codex · Cursor

A multiplayer workspace.

Your agents are fast. Your team is the multiplier.Open anyone’s running session in one click. Nothing to set up.

Talk to us
Also on
quicklist/quicklist
Search or run a commandK
0
JOMA
SASam Aydin
My queues
Workspaces
SAsam@quicklist.appquicklist · owner
in progress
fix/archive-completed-todos
The weekly digest reports zero completed todos for everyone. Find out why and fix it.
readsrc/todos/complete.ts0.2s
grep"completedAt" — 0 matches in src/0.3s
Found it. `completeTodo` runs a DELETE. The digest job counts rows where `completed_at` falls inside the week, and that column does not exist — so the count is zero by construction, for every account, every week.
editmigrations/0007_todo_completed_at.sql · +110.6s
editsrc/todos/complete.ts · +14 −60.8s
bashpnpm test todos digest6.4s
Completing a todo now stamps `completed_at` instead of deleting the row, and the list filters on it so nothing looks different to the user. Digest counts the week correctly. One call left for you: keep the history on the todos table, or move completions to their own table? The column is one migration; a separate table is a join on every read, forever.
claude opus 5 · padar thread
opus 5 · high
QL-118open in linear
Completed todos never reach the weekly digest
in-progressfrom Linearupdated 9m ago

The Monday digest counts everything a person finished that week. It has reported zero for three weeks running, for every account.

Ticking a todo off removes the row. By the time the digest job runs on Monday there is nothing left to count.

Done when · 0/3
Completing a todo keeps the record
The digest counts the week’s completions
The list still hides what you finished
Decisions · 0None yet. Mark a message in the thread as a decision and it lands here — and on the ticket.
quicklist/quicklist
Search or run a commandK
1
SAMA
JOJohn Okafor
My queues
Workspaces
JOjohn@quicklist.appquicklist · member
in progress
@sam tagged you· column or its own table?
fix/archive-completed-todos
The weekly digest reports zero completed todos for everyone. Find out why and fix it.
readsrc/todos/complete.ts0.2s
grep"completedAt" — 0 matches in src/0.3s
Found it. `completeTodo` runs a DELETE. The digest job counts rows where `completed_at` falls inside the week, and that column does not exist — so the count is zero by construction, for every account, every week.
editmigrations/0007_todo_completed_at.sql · +110.6s
editsrc/todos/complete.ts · +14 −60.8s
bashpnpm test todos digest6.4s
Completing a todo now stamps `completed_at` instead of deleting the row, and the list filters on it so nothing looks different to the user. Digest counts the week correctly. One call left for you: keep the history on the todos table, or move completions to their own table? The column is one migration; a separate table is a join on every read, forever.
SA@sam tagged you · column or its own table?now
claude opus 5 · padar thread
opus 5 · high
QL-118open in linear
Completed todos never reach the weekly digest
in-progressfrom Linearupdated 9m ago

The Monday digest counts everything a person finished that week. It has reported zero for three weeks running, for every account.

Ticking a todo off removes the row. By the time the digest job runs on Monday there is nothing left to count.

Done when · 0/3
Completing a todo keeps the record
The digest counts the week’s completions
The list still hides what you finished
Decisions · 0None yet. Mark a message in the thread as a decision and it lands here — and on the ticket.
quicklist/quicklist
Search or run a commandK
0
JOMA
SASam Aydin
My queues
Workspaces
SAsam@quicklist.appquicklist · owner
in progress
fix/archive-completed-todos
The weekly digest reports zero completed todos for everyone. Find out why and fix it.
readsrc/todos/complete.ts0.2s
grep"completedAt" — 0 matches in src/0.3s
Found it. `completeTodo` runs a DELETE. The digest job counts rows where `completed_at` falls inside the week, and that column does not exist — so the count is zero by construction, for every account, every week.
editmigrations/0007_todo_completed_at.sql · +110.6s
editsrc/todos/complete.ts · +14 −60.8s
bashpnpm test todos digest6.4s
Completing a todo now stamps `completed_at` instead of deleting the row, and the list filters on it so nothing looks different to the user. Digest counts the week correctly. One call left for you: keep the history on the todos table, or move completions to their own table? The column is one migration; a separate table is a join on every read, forever.
claude opus 5 · padar thread
opus 5 · high
QL-118open in linear
Completed todos never reach the weekly digest
in-progressfrom Linearupdated 9m ago

The Monday digest counts everything a person finished that week. It has reported zero for three weeks running, for every account.

Ticking a todo off removes the row. By the time the digest job runs on Monday there is nothing left to count.

Done when · 0/3
Completing a todo keeps the record
The digest counts the week’s completions
The list still hides what you finished
Decisions · 0None yet. Mark a message in the thread as a decision and it lands here — and on the ticket.
quicklist/quicklist
Search or run a commandK
0
JOMA
SASam Aydin
My queues
Workspaces
SAsam@quicklist.appquicklist · owner
in progress
fix/archive-completed-todos
The weekly digest reports zero completed todos for everyone. Find out why and fix it.
readsrc/todos/complete.ts0.2s
grep"completedAt" — 0 matches in src/0.3s
Found it. `completeTodo` runs a DELETE. The digest job counts rows where `completed_at` falls inside the week, and that column does not exist — so the count is zero by construction, for every account, every week.
readsrc/digest/weekly.ts0.2s
readprisma/schema.prisma0.3s
grep"DELETE FROM todos" — 2 matches0.2s
Both call sites go through `completeTodo`, so there is one place to change. The list query in `src/todos/list.ts` filters on nothing today — it just reads every row the table still has, which is why deleting was enough to make a todo disappear from it.
readsrc/todos/list.ts0.2s
The fix is a `completed_at` column and a filter on the list query — one migration and about fifteen lines.
writemigrations/0007_todo_completed_at.sql · +110.4s
editsrc/todos/complete.ts, src/todos/list.ts · +14 −60.9s
bashjust test digest — 34 passed6.1s
Migration applied against this workspace’s Postgres and the suite is green. Completing a todo now stamps `completed_at` instead of deleting the row, the list query filters on it, and the weekly digest counts the week’s completions again.
Ask the agent… @ for context, / for commandsopus 5 · high
Machine8 vCPU · 16.0 GiB
mem4.4 GiB
cpu1.86 load
.env files, against the project's
Environment.berth/Dockerfile
baked · bake-20260807T0914Z
Processesactions.toml
0 / 5 up
:8787 :5173
Servicescompose.yaml
postgres
redis
Actionsevery 5sactions.toml
stopped
Run onceactions.toml
not run
not run
Scriptsterminalactions.toml
devlive output
test digestlive output
Dataactions.toml
In this workspace
JOJohn Okaforhere
MHMarcus Haleidle

Spin up a workspace

Start fromquicklist
Branch
Issue2 open
Where
LocalRemote VM
Size
09:41
meridian
In Progress
booking-idempotencyfix/booking-idempotencyToday 09:38
receipt-emailsfeat/receipt-emailsToday 08:02
In Review
waitlist-smsfeat/waitlist-smsToday 07:41
table-mergefix/table-merge-capacityYesterday 18:55
Todo
seat-map-zoomfix/seat-map-zoomYesterday 17:24
no-show-feesfeat/no-show-feesMon 14:09
The one call it will not make alone reaches you wherever you are

The difference

Most agent tools still assume one person, one device.Padar assumes a team.

A single-player ADEPadar
Built around

Your session

The outcome

The bottleneck

How fast you write code

How fast your team takes the right decisions

A second person

A support request

A participant

Handing over

A description of the work

The work itself

The speed-up

Only you, until you need input from someone else

The whole team delivers faster

Isolation

Every workspace is a micro-VM.

  • You define your environment in Docker; Padar helps you set it up.
  • Each project securely shares a Docker layer with all of the images and volumes already extracted.
  • A new VM spins up in 2 seconds and instantly attaches to your project-wide containers and volumes.
  • Any changes you make to the containers and volumes stay local to your workspace and only persist in the shared layer if you decide so.
Runtimeper workspace
you16 GB
  • ·postgres:16
  • ·redis:7
  • ~pgdata
  • ·api:build-7
  • +ffmpeg
sam8 GB
  • ·postgres:16
  • ·redis:7
  • ·pgdata
  • ·api:build-7
  • +playwright
kit4 GB
  • ·postgres:16
  • −redis:7
  • ·pgdata
  • ·api:build-7
  • +seed.sql
shared layer
postgres:16redis:7pgdataapi:build-7

Latency

The environment feels local.The connection is end to end.

Tailscale opens a direct connection between your client and the VM. Nothing relays through us, which is why a remote shell types like a local one.

Your clientOne hopThe workspace

Whether you open your teammate’s laptop or a VM in the cluster, the connection is the same.

One button forwards a port out of the micro-VM. Click it and localhost:3000 is the app running inside, open in your browser.

Cross-functional by default

It starts on the ticket.Not in a setup doc.

A workspace that runs on a machine can be handed to anyone — a reviewer, a designer, the PM who wrote the ticket. There is nothing on their side to install.

Pick it in Linear

The workspace comes up on the right branch with the app already running. Nobody reads a README first.

Hand it over as a link

Reviewing means clicking through the running app, not reading a diff and hoping. No clone, no install, no “works on my machine”.

Take it with you

Prompt the agent, watch the app, approve a risky step — from iOS or Android. The same workspace, not a notification about it.

Control

Runs where you say it runs.

Nothing here is a lock-in you discover later.

Bring the model you already pay for

Claude Code, Codex or Cursor, signed in as you. Your subscription, your limits — nothing is resold through us.

Our cloud, or entirely your own

Run the machines on our infrastructure, or enroll your own and keep every workspace, every repository and every key inside your network.

Three hours of work.Two days to ship.

Agents made writing code fast. Deciding is what’s left.Padar puts the whole team in the same workspace. Upgrade your collaboration now.

Also on
Or talk to us