---

name: northcua description: Install, configure, run and troubleshoot the NorthCUA Desktop/Web Harness distributed by NorthCUA, including local controller workers, small browser batches, recovery and task evidence. Use when operating this repository; not for its unreleased training or data-generation systems.

Operate NorthCUA from a clean checkout

This is a reference snapshot bundled with the website. For execution, use the skill and linked guides from the separate NorthCUA Harness repository.

The project is NorthCUA, the shipped component is NorthCUA Harness, and the package/CLI is northcua. This skill is distributed with the repository. Resolve the real location of this file (including symlinks); the repository root is two directories above its folder. Run commands there, or use an installed release and the corresponding version of these guides. Do not copy this file alone: its linked guides are part of the skill's operating contract.

Choose the user's workflow

• Nothing installed / first task: follow from-zero setup. It includes exact installation commands, model templates, native permissions, a credential-free browser/native check and success criteria.
• Model or parameter configuration: configuration. config init --model ... --base-url ... populates both agents. The user normally supplies only model, key and optional endpoint via the env template. Keep credentials as environment references; do not ask for keys in chat.
• Several local tasks / recovery: follow local batch end to end. It starts a real controller, real browser workers and the existing scheduler without cloud services. Then consult batch operations.
• Python, serving, remote device: execution modes. Local application WebSocket and controller submission are different protocols.
• Custom runtime skill: skill loading.
• Failure, missing browser, model error or stuck task: troubleshooting.
• Result inspection / evaluation: evaluation.

Read the chosen tutorial before executing; it contains dependencies and expected outputs. Do not infer command flags from another CUA project or old experiment scripts. Use northcua <group> --help for optional controls and config schema for current app/batch fields. config validate does not call the model.

Execution contract

1. Determine the requested target (Web or native Desktop), execution mode, model endpoint and success criterion. Use the local scripted demo when checking the installation without credentials; label it as such. Use the user's real model for their actual task. Do not substitute demo success for task success.
2. Install from this repository's lock or matching release wheels. Activate the environment in every terminal. uv run --no-sync preserves installed extras. Create configuration in a new writable directory; preserve existing outputs.
3. Run one bounded task first. Desktop GUI tools return screenshots directly; there is no alternative observation mode. A physical display is exclusive: do not run concurrent Desktop tasks or use another agent's GUI session.
4. Inspect terminal status, the actual outcome, saved trace and evaluator verdict separately. A passing configuration check or actor completion is not outcome verification. Retain failed evidence and report unverified modes honestly.
5. For local batching use local-process, real browser workers and the existing batch commands. The fake provider is only a scheduling simulation. Start with concurrency 1–2; keep task IDs/output paths stable for resume. Drain and wait for the current runner to exit before resuming. Do not hand-edit state DBs.
6. Stop the processes you started and confirm cleanup. Preserve results for the user. The local provider shares the host OS and network; it is not a VM.

The self-contained path needs no NorthJetty, cloud key, private dataset or historical deployment checkout. E2B/OSWorld are optional, separate deployments; do not provision them unless that is the requested workflow.

Report the exact commands/mode, source revision, model type (scripted or real), actual outcomes, evidence paths, recovery/cleanup result and any remaining blocker. For an independent acceptance run, use a fresh output directory and follow these instructions without relying on author-provided success claims.