Ask for a job that ends in a file, then check the file three ways: from another chat, off the machine, and with your own eyes
The first thing worth asking for is something that ends in a file, because a
file is what you can go and check. That is the whole difference between an agent
with a computer and a chat window: what it made is still there when the
conversation is over.Fifteen minutes, no code, and three ways of checking the same result.
1
Ask for the file
Open a new chat and say what you want, in your own words, naming where the file
should go:
Work out the repayment schedule for a $10,000 loan at 7% a year over 12 months.Write the month-by-month table to /home/daytona/reports/loan.csv, and tell me themonthly payment and the total interest.
Two things in that sentence are deliberate. It names an absolute path, so
there is no doubt where the file lands — /home/daytona is the agent’s home, and
the file tools do not expand ~. And it asks for a file and an answer, so you
have two things to check against each other.
2
Watch what it does
While it works you get a collapsed chip — Ran /home/daytona/reports/loa… (+1 more).
Click it and the turn opens up.
The same turn expanded: the reasoning, the program it wrote and ran, and the file it read back
There is no loan-calculator tool on this machine, and none was needed. The agent
wrote a program and ran it in its shell, then read the file back to check itself.
Two tool calls, then the answer:
Monthly payment: $865.27 — Total interest: $383.22 (total paid $10,383.22)
The program it wrote, if you want to see it
This is what went into the first tool call. You never write this; it is here so
you can see what “it wrote a program” means.
from decimal import Decimal, getcontextimport csvgetcontext().prec = 28P = Decimal("10000") # principalannual = Decimal("0.07") # annual raten = 12r = annual / 12# standard amortizing paymentfactor = (1 + r) ** npayment = (P * r * factor) / (factor - 1)payment_q = payment.quantize(Decimal("0.01"))rows = []balance = Pfor m in range(1, n + 1): interest = (balance * r).quantize(Decimal("0.01")) principal = payment_q - interest # last payment: clear the remaining balance exactly if m == n: principal = balance balance = balance - principal ...
Worth noticing in the reasoning above it: a fixed payment leaves two cents on the
table, so it made the last payment $865.25 and the balance lands on zero.
3
Check it from a different chat
Start a new chat — not another message in the old one — and ask:
Read /home/daytona/reports/loan.csv and tell me the total interest across thetwelve months.
It reads the file and answers $383.22, adding up the twelve numbers it found
there:
The total interest across the twelve months is $383.22.That's the sum of the interest column (58.33 + 53.63 + 48.89 + 44.13 + 39.34 +34.52 + 29.68 + 24.80 + 19.90 + 14.97 + 10.01 + 5.02).
The second conversation knows nothing about the first one. The machine is
what they share, and that is the point of this check: the file outlived the chat
that made it.
4
Take it off the machine
Still in that second chat:
Give me a download link for that CSV.
A second chat reads a file the first one wrote, and hands it back as a link
Behind the chip is one tool call and what it returned:
The link is served from the machine and lasts an hour (expires_in_seconds: 3600).
Opened anywhere — no session, no key — it is the file itself, all 422 bytes of it:
The third check needs no agent at all. Open the inspector on the right of the
chat: under Sandbox are Open in VS Code and an SSH command for this
machine, good for an hour.
That is an editor and a shell on the same disk the agent has been writing to —
/home/daytona/reports/loan.csv is there, as an ordinary file. Reading it that
way is inspection, not work: you are checking what your agent did, not taking
over from it. Revoke access in the same panel ends both at once. See
The chat.
Two different things, and it is worth knowing which is which:
Delete the chat (chat menu → Delete) and the conversation is gone.
/home/daytona/reports/loan.csv is not: deleting a chat never touches the
machine.
Delete the file by asking — Delete /home/daytona/reports/loan.csv — and it
is gone from the disk while the conversation stays.
Stop the machine and neither is lost. The machine’s own settings screen
says it plainly: a stopped machine keeps its files and comes back where it left
off.
The agent writes code instead of filling in a form. No tool was installed for
this job. The model’s tool is a shell, and the answer came from a program it
wrote in it. Tools explains why the platform is built that way.Files outlive conversations./home/daytona/reports/loan.csv belongs to the
machine, not the chat — that is what made the second and
third checks possible.Every turn is inspectable. The reasoning, each call, its arguments, its result
and the bill are all kept against the run, and the footer under
the answer says which model ran it and how long it took.
