Builds and Runs

Understand how Workers run on the CoreClaw platform.

No Build Required — Lightweight Script Mode

CoreClaw’s execution flow is:

User Code → Script Runtime → Remote Fingerprint Browser Pool (CDP) → Target Website

CoreClaw uses lightweight process isolation, not container isolation:

• Runtime is pre-provisioned: Python/Node runtimes and base dependencies are already installed in the shared environment.
• Browser is remotely hosted: No need to build a browser environment locally. Instead, connect to the remote fingerprint browser pool via CDP.
• Dependency installation is automatic: The platform automatically installs dependencies based on requirements.txt / package.json — no manual image building required.

So all you need to do is: Upload your script (or find one in the Store) → Click Run. The platform handles dependency installation and execution automatically.

In short: CoreClaw has no Build step because it provides platform-level hosting for both the runtime and the browser. Users don’t need to worry about environment setup — this is CoreClaw’s advantage: upload and run.

Run Lifecycle

READY → RUNNING → SUCCEEDED / FAILED / ABORTING

Status	Description
READY	Run is queued and waiting to start
RUNNING	Script is currently executing
SUCCEEDED	Run completed successfully
FAILED	Run encountered an error
ABORTING	Run is being stopped

Run Management

• Monitor - View real-time run progress
• Abort - Stop a running Worker
• Rerun - Execute again with the same parameters
• View Logs - Debug execution issues

Run History

CoreClaw retains all your Worker run records. Each run saves:

• Unique Run ID
• Timestamp
• Duration
• Status
• Input parameters
• Output data
• Logs

You can view run records in the Console.

How many runs at once?

The number of runs you can execute at the same time depends on your plan. See Concurrency Limits.