No. It writes metrics to a local file for node_exporter to collect, and it can append blocked incidents to a local log file.
Yes. The plugin reads Wordfence data from the WordPress database. If required Wordfence tables or columns are unavailable, the exporter writes failure-state metrics instead of silently doing nothing.
The plugin schedules fast exports with WP-Cron. The default fast interval is every 15 minutes, and the admin UI also supports every 5 minutes, every 30 minutes, and hourly. Slow posture and scan metrics refresh hourly by default and can be set to hourly, twice daily, or daily. On low-traffic sites, WP-Cron may not run exactly on schedule unless you trigger WordPress cron processing through a system cron job or WP-CLI.
With the default metric prefix of wordpress_wordfence, the plugin can export:
It appends newly observed blocked Wordfence hits to a local .log or .jsonl path. The default text format preserves the original plain-text log line. The JSON Lines format emits one structured JSON object per blocked event for Loki, ELK, OpenSearch, and similar tooling. The exported incident timestamp is taken from the Wordfence hit row, falling back across known timestamp columns before using export time. The exporter tracks the last processed hit ID, and you can reset the incident cursor from the admin UI or WP-CLI to backfill retained history up to the configured per-run limit. For Loki, configure your log collector to parse the text prefix or the JSON Lines timestamp field if you want Grafana to display the original Wordfence event time instead of the collector ingestion time. Incident privacy controls can keep full IPs, truncate IPv4 to /24 and IPv6 to /64, hash IPs with the site salt, drop IP fields, drop query strings from URL and referer fields, drop referers, drop user agents, skip private/internal source IP ranges, and append an optional retention note to emitted events.
If WP-CLI is available, the plugin registers:
Yes. The source repository provides repository-only examples under examples/grafana/ and examples/prometheus/. They are intentionally not included in the WordPress.org plugin zip. The dashboard includes exporter health, activity, scan posture, WordPress version, plugin posture, opt-in plugin inventory, opt-in admin inventory, administrator login-protection coverage, and incident logs. Inventory-based alert examples require the matching opt-in inventory metric to be enabled.
The directory that will contain the .prom file must already exist and be writable by the PHP process running WordPress. If incident export is enabled, the incident log directory must also already exist and be writable by PHP. node_exporter must be able to read the resulting .prom file.