Repository navigation
Docs, slides and sheets: 3 tasks #116
Description
Test how well a UseAgent agent makes the files people use at work (documents, slides and spreadsheets) on three tasks, and report what happened.
Tasks
- A slide deck. Ask for a 6 to 10 slide deck on a topic you know, then ask for two rounds of changes.
- A document. A one to two page brief, plan or how-to, with headings, and a table where it fits.
- A spreadsheet that works. A budget, tracker or calculator with real formulas (totals, averages, a lookup), not just typed-in numbers.
Free to try
Sign up free at app.useagent.org. New accounts get free model credits (GLM or Claude Opus, up to a limit) and free cloud sandbox time, enough to finish this task without paying or adding a key.
- Sign up with your email at the link above. The link in the email works for 24 hours; check spam if it's missing. No API key, paid plan or card needed.
- Type the task in plain words and press send. The agent gets its own computer with a browser, terminal and desktop, and you watch it work.
- When it's done, copy the link from the address bar (
app.useagent.org/session/...). That link goes in your report.
Stuck? Comment here with a screenshot and we'll help.
How to do the runs
- A task is 10 runs. Your report has one line per run.
- Start each run yourself in the app, one at a time, and watch it work. Never more than 2 runs going at once. No scripts, tools or other AIs sending prompts for you.
- Every run needs at least one follow-up that reacts to that run's actual result: ask for a specific change and check it landed, or point out a specific problem. One run without this rejects the whole report.
- Record the whole session in one continuous, real-time screen recording that shows you starting each run and sending each follow-up. No cuts, no speed-ups.
Your report
Post one comment on this issue, one line per run:
Task: <number>
1. <thread link> - follow-up: <what you asked> - did it work: yes / no - <one sentence on what happened>
2. ...
...
10. ...
Recording: <link to the one continuous recording of your whole session>
- Each line describes its own thread. We read every thread before paying. Lines that don't match their threads, or repeat each other, reject the whole report.
- Reports that look batch-run or scripted aren't paid.
- Runs that didn't work still count, as long as you followed up on them.
- Before paying, we may ask about your own threads (for example, a price or a heading on the page the agent built). Answer from your threads.
- For the level of detail we expect, see the accepted report on #112.
Claim it
Comment "Taking task N". You hold the task for 3 days, then it opens up again.
Payment
- $5 per task, to the first person whose report checks out.
- One report per person per task. The first report that passes the checks gets that task.
- To get paid, email hi@useagent.org from the same email you signed up to the app with, and include a link to your report.
Keep private details out of here
This page is public. Don't post your real email, phone number, address, payment details, API keys, or screenshots that show any of them. Send anything private, including how you'd like to be paid, to hi@useagent.org with a link to your comment here.
Rules
Any of these means no payment, and we hide the comment without replying:
- Made-up runs or links, runs copied from someone else, or the same prompt ten times with tiny changes.
- More than one account per person, or a report for a task someone else is holding.
- Runs done or simulated by another AI tool. Do the runs yourself.
- Spam or harm: no mass form submissions, no scraping behind logins you don't own, nothing illegal.
Activity
[-]Docs, slides and sheets: 3 tasks, $5 each[/-][+]Docs, slides and sheets: 3 tasks[/+]joaorgb15 commentedon Oct 26, 2026
Taking task 2.
I'd like to confirm whether the $5 bounty is still active and available for a new submission. I previously completed and received payment for task 2 in #117, and I'd like to participate in this different task following all the testing and recording requirements.
Thank you!
joaorgb15 commentedon Oct 27, 2026
Task: 2 — Documents
https://app.useagent.org/session/4def06a8-6059-41e9-aff0-4b64562227b8 - follow-up: Remove unsupported R$500 and 48-hour limits, add responsible person and inspection date fields, and update the document. - did it work: yes - The agent revised the stock-control document and incorporated the requested corrections.
https://app.useagent.org/session/0dcee4bc-56ad-4c93-9bf9-e600b91edf73 - follow-up: Improve the responses for angry customers and online price comparisons without promising discounts. - did it work: yes - The customer-service manual was updated with the requested guidance.
https://app.useagent.org/session/f6c68342-047a-4840-8d47-2d0550d22c70 - follow-up: Remove the unsupported 20% revenue-growth target, make the goal adjustable, and generate a PDF. - did it work: yes - The marketing plan was revised as requested.
https://app.useagent.org/session/4bc3247d-da3b-47e0-8007-3604537dafdf - follow-up: Remove the fixed 08:00–18:00 schedule and add a supervisor training checklist. - did it work: yes - The employee-training guide was updated.
https://app.useagent.org/session/8d23b287-7288-4508-89bd-42d6a93bba40 - follow-up: Use FRAGILE labels only when necessary and add an order-number field to the checklist. - did it work: yes - The packaging procedure was corrected.
https://app.useagent.org/session/acf437f9-7e63-4874-9074-70e773ddeb1f - follow-up: Remove fixed 24-hour deadlines and add a status column to the meeting-minutes table. - did it work: yes - The meeting guide was revised with flexible deadlines and improved tracking.
https://app.useagent.org/session/d133196d-a679-403e-bec9-f687317e99f0 - follow-up: Remove unsupported sampling percentages and defect thresholds, and add batch and supplier fields. - did it work: yes - The quality-control document was updated.
https://app.useagent.org/session/5551e9e8-fc37-492f-b4d4-796df4f1a025 - follow-up: Add planned-versus-actual budget comparisons, identify fictional values and provide updated document formats. - did it work: yes - The financial-planning document was revised.
https://app.useagent.org/session/27e4436d-ec5b-4cfd-b481-1d6db8e5764f - follow-up: Add supplier delivery-risk information and a recommendation based on price, delivery time and quality. - did it work: yes - The supplier-comparison document was updated.
https://app.useagent.org/session/d1e89460-391a-48c0-97d6-c773093bcf6a - follow-up: Add a progress-status column and instructions for weekly review of operational delays. - did it work: yes - The improvement plan was revised with progress tracking and review instructions.
Recording: https://drive.google.com/file/d/1mYTyRRAdRiMEEgiIIQHb1_FkopwFrJ-e/view?usp=drivesdk
Recording disclosure:
During the initial attempt at run 8, UseAgent reached its usage limit. I paused the OBS recording and later resumed testing using the Codex connection available in UseAgent.
The original recording was preserved without editing out the interruption. I am disclosing this because I understand the task requires a continuous, unedited recording.
All 10 tasks and their follow-ups were performed manually, and I checked the resulting changes. I understand the recording interruption may affect eligibility and would appreciate confirmation of whether the submitted evidence can be accepted.
Thank you for reviewing my submission.
joaorgb15 commentedon Oct 30, 2026
Taking task 3.
joaorgb15 commentedon Oct 30, 2026
Task: 3 — Spreadsheets
https://app.useagent.org/session/bb663318-bf1a-40e1-bec9-4b347303fd34 - follow-up: Add formula-based replenishment quantities and costs, including totals, while preserving existing formulas. - did it work: yes - The updated output showed 7 units to replenish at a total cost of BRL 1,180.00.
https://app.useagent.org/session/1e723c44-bf82-454e-90ba-a273997cae1c - follow-up: Add a fixed BRL 2 fee per sale and update profit formulas and marketplace summaries. - did it work: yes - Reported total profit decreased from BRL 482.31 to BRL 470.31.
https://app.useagent.org/session/96243930-aabc-4975-be05-84ab3d5c6aa4 - follow-up: Change the exchange rate from BRL 5.40 to BRL 5.00 and recalculate import costs while retaining lookup formulas. - did it work: yes - Reported total import costs recalculated to BRL 29,000.00.
https://app.useagent.org/session/b166d512-beb8-42b4-b89a-58daf6fd6631 - follow-up: Add a fictitious BRL 1,000 December expense and recalculate monthly and accumulated balances. - did it work: yes - Reported year-end balance decreased from BRL 6,310.00 to BRL 5,310.00.
https://app.useagent.org/session/45742308-b348-4d98-a150-01e07541d255 - follow-up: Set the returned quantity for order PED-1001 to zero and recalculate return rates and refunds. - did it work: yes - The displayed total refund decreased from BRL 2,115.00 to BRL 1,935.20 and average return rate from 23.1% to 20.6%.
https://app.useagent.org/session/dcfc039c-c9ec-4e12-bf29-581ac3db9dc6 - follow-up: Add an editable BRL 100 bonus for salespeople reaching at least 110% of target and update total compensation. - did it work: yes - Two salespeople qualified, increasing reported total compensation from BRL 7,998.50 to BRL 8,198.50.
https://app.useagent.org/session/36ee0c98-f6f6-4024-b7ec-23ecb18c6cd7 - follow-up: Change the standard sales fee from 12% to 15%, preserving costs and margins and recalculating profits. - did it work: yes - Displayed average profit decreased from BRL 15.48 to BRL 11.01, with product EL-006 showing an increased loss.
https://app.useagent.org/session/30ddf579-bbfb-4d20-ad10-0e697afd4b0b - follow-up: Add an unpaid BRL 1,500 expense due October 20, 2026, using October 30 as the reference date, and recalculate overdue totals. - did it work: yes - Despite intermediate tool errors, the final output showed the expense 10 days overdue and total overdue amounts of BRL 2,760.00.
https://app.useagent.org/session/8a791932-b185-4079-aa01-65b74fabb361 - follow-up: Change supplier scoring weights to 80% cost, 10% delivery time and 10% quality, then recalculate the ranking. - did it work: yes - Supplier F002 became the top recommendation with a reported score of 96.25.
https://app.useagent.org/session/fdff4809-39e7-49e0-b55e-ac16756b7e27 - follow-up: Increase total units sold from 566 to 600 by adjusting only one product, then recalculate the dashboard while preserving formulas. - did it work: yes - Overall target achievement increased from 94.3% to 100%, and the agent reported preserving all 95 formulas.
Recording: https://drive.google.com/file/d/1TvwWk0lujJGh3czJMjjNuGhVWtyGakBt/view?usp=drivesdk