Execute tool calls in your customer's cloud. Your control plane sends commands, the worker runs them inside the customer's environment -- reading files, writing results -- without any data leaving their network.
Your Cloud Customer's Cloud
+-----------------+ +----------------------+
| AI Agent | -- command --> | worker |
| (reasoning) | <-- result -- | (this template) |
+-----------------+ | |
| +-- files --------+ |
| | Private storage | |
| +-----------------+ |
+----------------------+
| Resource | Type | Description |
|---|---|---|
worker |
Function (live) | Serverless function with command handlers |
files |
Storage (frozen) | Private file storage per customer (S3 / Cloud Storage / Blob Storage) |
| Command | Description |
|---|---|
execute-tool |
Run a tool by name (read-file, write-file) |
list-tools |
List all available tools and their descriptions |
alien devEverything runs locally -- storage on the filesystem, no cloud credentials needed.
In a second terminal:
# List available tools
alien dev commands invoke --deployment default --command list-tools
# Write a file
alien dev commands invoke \
--deployment default \
--command execute-tool \
--params '{"tool": "write-file", "params": {"path": "hello.txt", "content": "Hello!"}}'
# Read it back
alien dev commands invoke \
--deployment default \
--command execute-tool \
--params '{"tool": "read-file", "params": {"path": "hello.txt"}}'alien dev deploy --name acme-corp --platform localEach deployment gets isolated storage -- files written by default are invisible to acme-corp.
Change your code, then:
alien dev releasebun testSee From Local to AWS to deploy this worker into a customer's AWS account.