autoscaling

A zero-dependency, in-process fake of AWS Auto Scaling (EC2 Auto Scaling groups, launch configurations, and launch templates). Speaks the AWS Query (XML) wire protocol (API version 2011-01-01, member-style lists), so the real @aws-sdk/client-auto-scaling works against it unchanged.


Port	`4706`
Protocol	AWS Query / XML (API version `2011-01-01`, member-style lists)
Health	`GET /_parlel/health`
Reset	`POST /_parlel/reset`

Default connection

AWS_ENDPOINT_URL=http://127.0.0.1:4706
AWS_ACCESS_KEY_ID=parlel
AWS_SECRET_ACCESS_KEY=parlel
AWS_REGION=us-east-1

Any SigV4 credentials are accepted (auth is not verified).

Implemented operations

Category	Operations
Auto Scaling groups	`CreateAutoScalingGroup`, `DescribeAutoScalingGroups`, `UpdateAutoScalingGroup`, `DeleteAutoScalingGroup`, `SetDesiredCapacity`
Launch configurations	`CreateLaunchConfiguration`, `DescribeLaunchConfigurations`
Launch templates	`CreateLaunchTemplate` (technically an EC2-API operation, included here for convenience)

Setting/updating DesiredCapacity synthesizes or removes placeholder instances (i-…) so describes reflect the requested capacity.

SDK usage example

import { AutoScalingClient, CreateAutoScalingGroupCommand, SetDesiredCapacityCommand } from "@aws-sdk/client-auto-scaling";

const asg = new AutoScalingClient({
  endpoint: "http://127.0.0.1:4706",
  region: "us-east-1",
  credentials: { accessKeyId: "parlel", secretAccessKey: "parlel" },
});

await asg.send(new CreateAutoScalingGroupCommand({
  AutoScalingGroupName: "web-asg",
  LaunchConfigurationName: "web-lc",
  MinSize: 1, MaxSize: 5, DesiredCapacity: 2,
  AvailabilityZones: ["us-east-1a"],
}));
await asg.send(new SetDesiredCapacityCommand({ AutoScalingGroupName: "web-asg", DesiredCapacity: 4 }));

Access via MCP / preview URL

When running inside a Daytona sandbox, this HTTP service is exposed at an automatically-provisioned preview URL. Point the SDK endpoint at that URL and add the x-daytona-preview-token header on requests.

Surface coverage

This emulator faithfully replicates the API surface most application code and agents exercise. Anything below the supported lines is either an intentional design choice for a fast, zero-cost local emulator (✓ By design) or a candidate for a future release (⟳ Roadmap) — never a silent inaccuracy.

Legend: ✅ fully supported · ◐ accepted (stored, not strictly enforced) · ✓ by design · ⟳ on the roadmap.

Area	Status	Notes
CreateAutoScalingGroup	✅	Full request parsing, required-field validation, duplicate name detection.
DescribeAutoScalingGroups	✅	Filter by name, all response fields including empty elements for optional fields.
UpdateAutoScalingGroup	✅	Partial update of min/max/desired/cooldown/health check/AZs.
DeleteAutoScalingGroup	✅	ResourceInUse guard, ForceDelete support.
SetDesiredCapacity	✅	Scales synthetic instances up/down to match desired capacity.
CreateLaunchConfiguration	✅	Full field set: ImageId, InstanceType, KeyName, SecurityGroups, UserData, EbsOptimized, InstanceMonitoring, etc.
DescribeLaunchConfigurations	✅	Filter by name, all response fields.
CreateLaunchTemplate	✅	Minimal single-version template; returns LaunchTemplateId (lt-…).
Error envelope	✅	Matches real AWS XML error shape: ErrorResponse > Error > Type/Code/Message + RequestId. Status codes match real API.
Response XML namespace	✅	`https://autoscaling.amazonaws.com/doc/2011-01-01/` (matches real API).
Empty element handling	✅	Optional empty fields rendered as self-closing `<tag/>` (matches real API).
Instances	✓ by design	Synthetic placeholders (`i-…`); nothing launched in the EC2 emulator.
Scaling policies	⟳ roadmap	Target tracking / step scaling policies & CloudWatch alarms are not implemented.
Lifecycle hooks	⟳ roadmap	Lifecycle hooks, warm pools, and instance refresh are not modeled.
Health checks	◐	`HealthCheckType` is recorded but no health evaluation happens.
Launch templates	◐	A minimal single-version template only; `CreateLaunchTemplateVersion` is not implemented.
MixedInstancesPolicy	⟳ roadmap	Not yet parsed or stored.
Auth	✓ by design	SigV4 is accepted but never validated.
Persistence	✓ by design	In-memory; lost on restart/reset.

Error codes & shapes

Errors follow the real AWS Auto Scaling XML error envelope:

<ErrorResponse xmlns="https://autoscaling.amazonaws.com/doc/2011-01-01/">
  <Error>
    <Type>Sender</Type>
    <Code>ValidationError</Code>
    <Message>Human-readable error message</Message>
  </Error>
  <RequestId>…</RequestId>
</ErrorResponse>

Code	HTTP Status	When
`ValidationError`	400	Missing required field, invalid action, invalid parameter.
`AlreadyExists`	400	Duplicate ASG or launch configuration name.
`InvalidParameterValue`	400	Reserved for future validation.
`InvalidParameterCombination`	400	Reserved for future validation.
`MissingParameter`	400	Reserved for future validation.
`ResourceInUse`	400	DeleteAutoScalingGroup with instances and no ForceDelete.
`InternalFailure`	500	Unhandled server error.

The <Type> element is Sender for 4xx errors and Receiver for 5xx errors, matching the real AWS behavior.

Manifest

{
  "name": "autoscaling",
  "version": "0.1",
  "port": 4706,
  "protocol": "http",
  "healthcheck": "/_parlel/health"
}

Configuration — `test.env`

Copy these into your test.env (used by the bridge sidecar flow). Tokens are Parlel's seeded test credentials — any non-empty value is accepted by the emulator, so you rarely need to change them. Swap in real credentials only when pointing at the live service in prod.env.

AWS_ACCESS_KEY_ID=parlel
AWS_SECRET_ACCESS_KEY=parlel
AWS_REGION=us-east-1
AWS_ENDPOINT_URL=http://parlel-bridge:4706