islo
DocsBlogSign inStart an agent →

Run untrusted code off your laptop.

Islo runs untrusted or LLM-generated code on disposable isolated computers instead of your laptop or production network.

Execute LLM-generated scripts on throwaway isolated computers with egress policies, then discard the machine when the job is done.

Build your first autonomous agent →Learn more
agent computer
cpu2 vCPU · 4 GB
disk10 GB SSD
staterunning
a real computer · wired to everything an agent needs
start withand more
onIslo cloudyour cloudyour customers' cloud
agent
fetch("https://pypi.org/simple/requests")
$PYPI_TOKENplaceholder only
islo gateway
Bearer $PYPI_TOKEN→Bearer ••••••••4f2a
injected at egress · per-request audit
pypi.org
200 OK
key never touched sandbox memory
allowpackage registrypypi.org onlyscoped
allowresult uploads3://results/*scoped
allowephemeral /tmpworkspaceaudited
agent-run-ab12 · isolation
python
agent script
/tmp
ephemeral
gateway
per-request policy
→ isolated · scoped egress · VM destroyed
computer
agent-run-ab12
isolated run · destroyed after
mounted at/workspace
storage volume
vol-ab12 · 2 GB
read/write for the run
readinputs/data.csv
writeresults/output.json

Pay only while it runs.

this use case30s run · 1 vCPU / 512 MB → ≈ $0.002

Pay only while the computer runs: CPU + memory + disk. No seat licenses, no idle fees.

CPU time
$0.07/CPU-hour
per core while the env runs
Memory
$0.04/GB-hour
per GB you provision
Storage
$0.0007/GB-hour
per GB of disk
$50 free credits on signup · no card required

From our blog

Agent infrastructure, in depth.

View all →

Common questions

Why shouldn't I run LLM-generated code on my laptop?

Generated scripts may be wrong or malicious. Islo runs them on disposable computers with scoped egress instead of your laptop or production network.

Can I send Python from an LLM response straight to Islo?

Yes. Pipe generated code into an isolated computer through the CLI or SDK and collect stdout, stderr, and artifacts.

Is the computer deleted after the script finishes?

You can stop and remove the computer, or snapshot state if you need to inspect results later.

Can the generated script access the internet?

Only through the egress rules you configure. Gateway policies can allow required hosts and block everything else without placing credentials inside the computer.

Can I inspect stdout, errors, and generated files afterward?

Yes. Collect stdout, stderr, and artifacts through the CLI or SDK, and keep or snapshot the computer when you need a longer investigation.

Put autonomous agents to work.

Build your first autonomous agent →how it works
© 2026 Islo
DocsBlogLLM contextAGENTS.mdTermsPrivacyRefundsSign in