Audit hosting

Read-only from start to finish. The output is a report; every fix is a separate, confirmed step through another skill.

Calling the operations

• Read the inputSchema that search returns before the first call of each operation. If a name below is rejected as unknown, search for what the step does (e.g. "ssl status").
• Batch reads with multi-execute, up to 20 steps a batch. Batches chain operations from one server only — the Hostinger Connector runs hosting, wordpress, agency-hosting and billing as separate servers. When a product group is switched off, skip its checks and say which were skipped.
• Page every list until meta.total is covered.
• Leave hosting_php_get out — it is large and says nothing about account health.

1. Inventory

One batch per server:

• hosting_orders_list with statuses: ["active", "suspended"]. cloud_* plan names are Cloud; the rest (hostinger_premium_*, hostinger_business_*, …) are Shared.
• hosting_websites_list — every Shared and Cloud website with website_type, username, order_id and is_enabled (false means suspended).
• wordpress_installations_list with ownership: "all" — id, is_valid, validation_error.
• agency-hosting_orders_list and agency-hosting_websites_list-plan — Agency orders and sites (details.uid, details.type, details.domains, details.state).

Tell the user the size (plans, websites, WordPress installs) before the checks. For large accounts offer to scope the audit to one order or domain.

2. Checks

Per Shared and Cloud website (skip builder and horizons):

• hosting_ssl_status — flag failed (with last_error), expired, not_installed on a custom domain, is_https_redirect_enabled: false, and expires_at within 30 days when is_lifetime is false. Lifetime certificates renew on their own.
• Node.js sites: hosting_nodejs_list-builds with per_page: 1 (latest build failed?) and hosting_nodejs_list-vulnerabilities with severities: ["critical", "high"].

Per hosting account (username): hosting_databases_list — flag disk_usage_mb above 80% of max_size_mb.

Per WordPress install: wordpress_installations_show-core-version (core vulnerabilities) and wordpress_plugins_list-installed (vulnerable plugins and pending updates).

Per Agency order: agency-hosting_metrics_list-order-resource-usage with time_frame_hours: 168 and agency-hosting_metrics_list-plan-order-disk-usage with time_frame_days: 7 — flag usage above 80% of the plan quota and name the heaviest websites.

Per Agency site: agency-hosting_ssl_website-status for each custom domain in details.domains; flag details.state other than active.

Plans about to lapse (billing product): billing_subscriptions_list, joined to hosting orders on subscription_id — flag not_renewing, or is_auto_renewed: false with expires_at within 30 days. Every site on that plan goes down with it.

From outside, for each custom domain:

curl -sS -o /dev/null -w "%{http_code}\n" --max-time 15 https://DOMAIN/
curl -sSI --max-time 15 https://DOMAIN/ | grep -i "^platform"

No platform: hostinger header means the domain does not reach this hosting.

3. Report

# Hosting audit
3 plans (2 Shared, 1 Cloud) + 1 Agency · 14 websites (6 WordPress, 3 Node.js, 5 other) · 4 Agency sites

## Fix now
- shop.example.com — latest Node.js build failed 2 days ago → troubleshoot-website
- blog.example.com — contact-form-x 5.2 has a known vulnerability, fixed in 5.3 → maintain-wordpress

## Soon
- example.org — Business plan expires in 12 days, auto-renew off → renew in hPanel
- Agency order 1000000001 — 86% of disk quota, mostly client-a.com

## Worth knowing
- 4 WordPress installs have pending plugin updates → maintain-wordpress

## Healthy
docs.example.com, portfolio.example.com, …

• Fix now: site down or not reaching Hostinger, failed latest build, invalid WordPress install, SSL failed or expired, critical vulnerabilities, suspended website.
• Soon: high vulnerabilities, a plan lapsing within 30 days, usage above 80% of a quota, HTTPS redirect off, non-lifetime certificates expiring.
• Worth knowing: pending updates, inactive plugins with vulnerabilities.

Each item names the evidence and the skill that fixes it: troubleshoot-website, maintain-wordpress, connect-domain or deploy-to-hosting. Patchable npm vulnerabilities on GitHub-deployed sites can be fixed with hosting_nodejs_patch-vulnerabilities, which opens a pull request — offer it, do not run it as part of the audit.

Not available through the API

CPU and memory usage for Shared and Cloud plans, website backup status, and PHP error logs. Mention them once at the end so the user knows what the audit could not see.