[wrkr]DocumentationGet help ↗
Workspace guide

What can an AI worker do?

Practical Wrkr workflows for research, documents, recurring work and small apps: what to provide, what to ask for and what to review.

Planned launch offering

Wrkr is preparing for invitation access. Paid checkout is not open. These guides describe the planned service; example requests are starting points you can adapt. See pricing and availability.

A Wrkr AI worker uses your chosen AI, files, browser and Linux tools to work on a task. Give it an outcome and relevant source material, then open the result and review it. The examples below are starting requests you can adapt.

Copy, edit or download a ready-to-use task brief →

Research a question and save the evidence

Provide: the question, scope, deadline and any sources you already trust. For a market brief, specify the market and date range instead of asking for “everything.”

EXAMPLE REQUEST

“Compare these three services for our five-person team. Read their official pricing and documentation. Save a comparison with source links, access dates, costs and unresolved questions. Do not sign up or buy anything.”

Review: open the sources, check dates and confirm that prices refer to the same billing period. A saved report is something you can inspect; it is not a guarantee that the AI found every relevant fact.

Turn working notes into a document

Provide: meeting notes or project files, the intended reader and the desired format.

EXAMPLE REQUEST

“Read the notes in this project. Save a one-page client update and a separate action list. Keep confirmed decisions separate from open questions. Flag missing owners or deadlines instead of inventing them.”

Review: open the saved files through Files, check names and decisions, and ask for revisions. Keep important project decisions in files so a fresh thread can continue the work.

Prepare a recurring update

Provide: accessible sources, a schedule with a time zone, an output folder and the decisions that need your attention.

EXAMPLE REQUEST

“Every Monday at 9 am Asia/Kolkata, review the project’s open tasks and save a weekly update. Include changes since the last saved update and questions for me. Prepare a draft; do not send it.”

Review: confirm the schedule and look for a saved result after a run. Work requires the computer to be running, available capacity and valid provider access. A sign-in or human decision can interrupt it. Outside services can incur charges while you are away; their bills are separate.

Build a small app and revise it

Provide: a small problem, a sample of the data and the audience. Start with a demo before putting real customer information into it.

EXAMPLE REQUEST

“Build a booking demo using sample data. Show available slots and prevent a slot from being booked twice. Add an owner view. Explain how to test it before sharing the worker’s app link.”

Review: test booking conflicts, access controls and saved data. Your AI builds and maintains the app; Wrkr supplies local capabilities within shared computer capacity. Read what app hosting includes before sharing it.

Choose one task you can check

Use a separate worker for a responsibility that needs its own browser sign-ins and files. Start with a small result, review it and refine the request. Create your first worker, or request an invitation when you are ready to explore Wrkr.