Representatives

Your work keeps answering after you log off.

Every person, team, and project on StandIn has a Representative. When you're heads-down, in a meeting, or offline for the day, your teammates can still ask the questions they'd normally interrupt you for — and get sourced answers from what you already wrote.

You stay in charge of your work. StandIn just extends your reach.

Sarah's Personal Representative · online
> Alex asks

Is the API migration blocked?

Yes. Blocked on LIN-482, waiting on Dave's review. Sarah flagged this in her 17:00 wrap.
Source: Sarah Chen · wrap published 17:02 GMT · 3h ago
01Three kinds

One idea, three scopes.

StandIn gives you three kinds of Representative. Each stands in for something that can't be interrupted — a person, a team, or a project — and all three follow the same rule: they only speak from what's already been written.

i.

Project Representative

Stands in for the work itself.

Projects are where strategy meets execution. They cross teams, cross time zones, and carry the decisions that shape whether you hit your goals.

Every wrap that touches the payments project becomes part of its Representative — the gateway vendor decision from two weeks ago, the architectural call Dave made Monday, the blocker Priya flagged this morning. All held together, searchable by anyone working on the project.

A new engineer joins and asks what she needs to know. She gets the answers in the voices of the people who made the decisions, with dates and reasoning attached. Onboarding stops being a scavenger hunt.

A product manager wants to know if the project is on track. She asks, and gets a sourced picture of what's shipped, what's blocked, and what's still in motion. No status meeting. No pulling six people into a Zoom.

The Project Representative is where business goals stop being a deck and start being a living picture of the work.

ii.

Team Representative

Stands in for a whole team.

A team's context doesn't live in any single person. It lives in the sum of their wraps — the reviews, the handoffs, the rituals, the small call-outs that add up to how this team actually works.

Ask the team's Representative who's on what, what the team decided this sprint, or how they handled something similar three months ago. You get answers drawn from the people doing the work, each one cited by name.

When one teammate is out, the team doesn't go dark. When someone leaves, the record stays.

iii.

Personal Representative

Stands in for one person.

Your Personal Representative answers in your voice, from what you've published. It doesn't guess, doesn't invent, doesn't speak for decisions you didn't make.

When you're in deep work, traveling, or just off the clock, teammates can still reach the things you've already said — without pinging you.

Quiet is the default. If the wrap doesn't cover the question, the Representative says so and points back at you — or at whoever can actually answer.

02The one rule

Three possible responses. No fourth option.

Every question a Representative hears resolves into one of three outcomes. There's no category for guessing.

A

Answer

When the wrap covers it, the Representative answers — and cites the wrap.
> “Is the API migration blocked?
Yes. Blocked by LIN-482, assigned to Dave.
Source: Sarah's wrap, 17:00 GMT
B

Redirect

When someone else wrote it, the Representative hands off.
> “What's the rollback plan for Friday?
Priya owns the rollback plan — see her wrap from Tuesday, or ask her directly.
C

Refuse

When no one has written it, the Representative declines cleanly.
> “What did the team decide about the Stripe integration?
No wrap has addressed that decision. I won't guess — ask @Sarah or @Liam.
03Not a chatbot

What it isn't.

A Representative isn't a chatbot. It isn't a summarizer. The difference is what each one is allowed to do when it doesn't know.

A chatbot
An AI summarizer
A Representative
Pretends to be the person
Paraphrases their voice
Quotes what they actually wrote
Generates responses from a model
Produces summaries that may be wrong
Cites every answer to a specific wrap
Has no accountability
Obscures the source
Names the author on every line
Invents when it doesn't know
Fills gaps silently
Refuses and names who to ask

The difference matters the first time a confident AI summary turns out to be wrong. Representatives are usable for real decisions because the honesty is real — you either get a sourced answer or you get silence.

04What it never reads

The things a Representative deliberately doesn't know.

A Representative's trustworthiness comes from what it refuses to look at, not just what it can see.

No private messages.

Representatives never read DMs, private Slack threads, or closed channels. Public, declared context only.

No calendar surveillance.

No tracking of when you're online, busy, or idle. Presence is something you publish, not something inferred from signals.

No activity monitoring.

No keystroke capture, no screen-time logs, no mouse heatmaps. A Representative knows what you wrote, not what you were doing when you wrote it.

This isn't a limitation. It's what makes the answers trustworthy. A Representative you can rely on is one that only speaks when the person it stands in for gave it permission to.

05Honest question

If it only knows what's in the wrap, isn't that limiting?

Yes. That's the point.

Imagine a tool that tells you the truth 90% of the time and invents something the other 10%. You spend more time verifying than you save. You can't defend a decision built on it, because you never know which 10% you're standing on.

Representatives are usable because the honesty is real. When you get an answer, you can act on it. When you get nothing, you know to ask the person directly. Either outcome moves the work forward. Neither produces debt.

See a Representative in action.