Built into Pencel — the screen you open first

Agentic Work Command Center

Home

Home is where every Program reports in. Work in progress and anything paused on your approval sit in Active; finished results land in Feed as briefing cards you read in a reading pane and act on without leaving; your own to-dos sit beside them. A run that found nothing stays quiet — recorded, but not published — so a fast cadence never turns this into another inbox to clear.

Every Program reports here

S
Scoutfiles what it found overnight
M
Mailroombrings the threads that need you
L
Ledgeranswers from the live books
3 scopes
Active, Feed and To-dos — everything waiting on you, one tab away
Silent
A run that found nothing publishes no card, so quiet stays quiet
In place
Ask and reply from the card itself, without opening a room

See it on Home

What it looks like with this running.

Tuesday · 7:58 AM

Home
ActiveFeedTo-dos
S

Scout6:04 AM

Filed overnight, waiting for you on Home.

3 opportunities cleared your fit bar

From 41 listings across SAM.gov and three state portals. Top one scored 87/100, due in 9 days.

Roll-up cardOpens Opportunities
M

Mailroom6:31 AM

Three threads need you. The rest are cleared with a reason.

Draft reply — Ortega Law, rate question

Written against your standing pricing rule. Review it here; sending goes through a gate.

Review and sendAsk a follow-up
L

Ledger7:15 AM

Month-end summary is up. No decision needed.

Two other runs finished overnight and found nothing. Recorded on Jobs, no cards published.

You

You7:58 AM

Add a to-do: chase the Port Authority reference before Friday.

Added to To-dos. Hand it to an agent with Run once, or build it into something recurring.

End to end

The whole loop, start to finish.

Every step below runs today. Nothing here is a mock-up of what is coming.

  1. 1LandEvery Program publishes what it finished here as a briefing card, with the agent, the headline and a few fact chips.
  2. 2ReadThe list reads like an inbox; the reading pane beside it gives you the highlights, the sources and why you are seeing this.
  3. 3ActAsk a follow-up or draft a reply from the card itself. Nothing external goes out without passing a review gate.
  4. 4DecideAnything paused on your approval or waiting for an answer sits in Active until you clear it, not buried in a room.
  5. 5DelegateAdd Task hands new work to an agent — once, right now, or compiled into something that runs on a schedule.

Capabilities

Everything it does, named.

One line each. No adjectives — every one is a specific thing you can go and watch it do.

  • Collects what every Program finished into one feed, so overnight work is one screen rather than a tour of rooms.
  • Splits the day into Active (running now, or paused on your approval), Feed (finished results) and To-dos (yours), so nothing waiting on you is more than one tab away.
  • Reads like an inbox: a compact card list on the left with agent, topic, headline and fact chips, and a reading pane on the right with highlights, sources and why you are seeing it.
  • Puts Ask on every card for a quick read-only follow-up to the agent that produced it, without opening its room.
  • Puts Reply on every card that has a source to reply to, and holds the draft behind a review gate so you approve the exact words before they leave.
  • Stays quiet when a run found nothing: the run is recorded on Jobs, no card is published, and a fifteen-minute cadence never floods the page.
  • Collapses a run that produced many records into a single roll-up card that routes you into its own scope tab, where the per-record actions live.
  • Adds a dedicated tab for a pack’s record type when you install one — Opportunities, for instance — and never shows it to anyone who has not.
  • Turns Add Task into either an ephemeral run that happens now or a compiled, scheduled job, from the same button.
  • Sorts and filters the list the way you want, with importance as one ordering option rather than the only one.

Inputs and outputs

What it reads, and what you get back.

What it connects to

Every Program you have installedYour to-dos, workspace-wideJobs — the run behind any cardGmail, Slack and anywhere else a reply would goRadar insight cards

What lands in your hands

  • One morning read of everything that ran while you were away
  • Decisions and approvals in one queue instead of scattered across rooms
  • A reply you approved, sent where it needed to go
  • A new task running, once or on a cadence

A week with it running

Day one to steady state.

What the first day looks like, what the quiet days look like, and when it comes to find you.

  1. Every morningOpen one screen and see what ran overnight, what it found, and what needs a decision from you.
  2. Through the dayResults land as they finish. Read, ask a follow-up, approve a draft, check off a to-do.
  3. When nothing happenedNothing appears. Quiet is a real state here, not an empty inbox you still have to check.
  4. When you want moreAdd Task hands the next thing to an agent, once or on a schedule.

Guardrails

Where it stops for you.

The line between what it does on its own and what it will never do without you, stated plainly.

Waits for you

Every approval and every input request, which sit in Active until you clear them, and every reply, which is held behind a review gate before it goes out.

Never does

Send a reply you have not approved, or hide a run from you — a run with nothing to say still records itself on Jobs.

Home is the first screen you see. Run history and schedules live on Jobs.See your first morning read

Get started

Three steps to your first real result.

  1. 1Download Pencel. Home is the screen you land on.
  2. 2Install one Program, or use Add Task to hand an agent something you do every week.
  3. 3Come back the next morning and read what it filed.

macOS, Windows and Linux. No invite code.

Read the docs first →