hub-pro-cloud-api

Operate Hub Pro cloud proceed, founder-ops API, and Railway deploy workflows via local hub endpoints.

Updated Jun 20, 2026
One-click install
npx skills add https://github.com/Noetfield-Systems/SourceA --skill hub-pro-cloud-api-noetfield-systems
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hub-pro-cloud-api
Source: https://github.com/Noetfield-Systems/SourceA/tree/main/.cursor/skills/hub-pro-cloud-api
Command: npx skills add https://github.com/Noetfield-Systems/SourceA --skill hub-pro-cloud-api-noetfield-systems

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It centralizes the operational knowledge for running Hub Pro cloud tasks—proceeding cloud Forge Run batches on Railway, calling the founder-ops API, and using the API Station—without running a local FORGE motor on the Mac founder session. ## Core Features & Use Cases - Cloud Proceed Control: Trigger full-pack cloud task proceeds through the Cloud Workers API on port 13027 with receipts and logs for verification. - Founder Ops API: Query and execute founder operations such as dual_heal and cloud worker status checks via the Worker Hub on port 13020. - Deploy & Queue Verification: Deploy to Railway with the founder deploy script and verify queue state through the cloud-forge-run queue endpoint, including stale pointer repair. - Use Case: After a Railway deploy shows healthy but the queue still references an old batch, use the documented sync_pointer_from_image repair flow to realign the persistent volume pointer and confirm the new batch. ## Quick Start Ask the assistant to proceed the next cloud task using the Cloud Workers API and show the receipt and queue proof.

Frequently Asked Questions about hub-pro-cloud-api

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I proceed the next cloud task in Hub Pro?▼

Send a POST request to http://127.0.0.1:13027/api/cloud-workers/v1 with the action set to proceed, full_pack true, and full_motor true. The receipt is written to ~/.sina/hub-cloud-forge-run-proceed-receipt-v1.json and the run is logged in the proceed log file.

How do I verify a Railway deploy actually updated the cloud queue?▼

Do not rely on the /health endpoint, which only proves the service is alive. Query the /api/cloud-forge-run/queue/v1 endpoint and check cloud_forge_run_head, observed.batch_id, and queue_batch_complete for the expected new batch.

Why does the Railway queue still show the old batch after a successful deploy?▼

The persistent Railway volume pointer can override the image pointer, leaving the queue on the old batch. POST to the queue endpoint with action sync_pointer_from_image to repair it, then re-read the queue to confirm.

What operations does the founder-ops API support?▼

The founder-ops API on port 13020 accepts GET and POST requests at /api/founder-ops/v1, supporting operations like dual_heal and cloud category checks including cloud_workers_status, probe, dry_run, deploy instructions, and proceed dry runs.

Can I run the full FORGE motor locally on the Mac founder session?▼

No. Full motor FORGE runs for Cloud Forge Run rows are forbidden on the Mac founder session. Proceeds must go through Cloud Workers.app or CF cron to Railway, and chat-only proceeds without a hub receipt are not valid.