Your WordPress site runs itself.

Every action is classified into three tiers. Read-only actions (analytics, scanning, monitoring) happen without approval. Low-risk changes (meta tags, schema updates, cache purges) deploy automatically with instant rollback. High-risk changes (new pages, plugin updates, structural edits) go through a 15-minute rollback window or require your approval. You control how much autonomy each agent has.

Every action your agents take is logged — which tasks succeeded, which failed, what data they found. Agents use this context in subsequent cycles to make better decisions. For example, Lina remembers which SEO fixes she already applied, Bastien tracks which files changed since the last integrity scan, and Hugo knows which leads he already scored. It’s cumulative context, not a generic playbook.

Through a shared event bus. When Selma detects a traffic spike from Reddit, she signals Lina with the source and intent data. Lina creates a targeted page and signals Hugo to monitor for leads. Every inter-agent message is logged and visible on your dashboard. If two agents try to modify the same resource, Thomas halts both and asks you to decide.

Before any modification, the agent creates a snapshot. After the change, it runs a health check within 60 seconds. If the check fails, the change is automatically rolled back. If two consecutive health checks fail, all agents halt write operations and alert you. In practice, most agent work is read-only — only about 15% of actions modify your site at all.

No. WP-Claw replaces operational plugins — SEO (Yoast), security (Wordfence), analytics (MonsterInsights), backup (UpdraftPlus), CRM, and live chat. It also monitors performance (Core Web Vitals, database optimization) but does NOT replace your caching plugin — keep WP Rocket or similar for caching. Your page builder, theme, and design tools are untouched.

No. WP-Claw works alongside existing plugins. Once you’re confident an agent handles a capability well, you can deactivate the equivalent plugin at your own pace. No forced migration.

All agent infrastructure runs on Hetzner in Germany (EU). Your WordPress data stays on your own hosting. Agents communicate via encrypted API — we don’t copy your database. API keys and secrets are encrypted at rest with AES-256-GCM. Agent task data is stored on EU servers.

Everything. Your dashboard shows a real-time activity log of every decision, every action, every inter-agent signal. You can see why an agent made a specific change, what data it used, and what the measured outcome was. Full transparency is non-negotiable.

Yes. Hugo handles abandoned cart recovery, lead scoring, inventory alerts, and customer segmentation. Selma monitors conversion funnels and signals Hugo when commercial traffic patterns appear. They work together on your WooCommerce store the way a growth team would.

A mix of Claude (Anthropic), Gemini (Google), and specialized models — routed through EU infrastructure via OpenRouter. Each agent uses the best model for its task type: fast models for monitoring, capable models for content. AI usage is covered by credit packages (from 25€/mo) or bring your own API keys.

No. WP-Claw includes built-in health monitoring at every layer. Agents check WordPress connectivity every 5 minutes with automatic circuit breakers and self-recovery. Daily LLM spending is capped per instance to prevent surprises. A live status page at wp-claw.ai/status pings actual endpoints every 30 seconds. No UptimeRobot, Betterstack, or any third-party service needed.

Founding members get a 30-day trial on any plan. After the trial, you’re billed at the founding member rate — locked for life. No credit card required to start.

Cancel anytime from your dashboard. Your subscription stays active until the end of the billing period. After that, agents stop working and you can export your data. No lock-in, no penalties.