Patchstack Security MCP. Audit vulnerabilities and software status via chat.

Patchstack Security lets your AI agent audit WordPress and PHP security across multiple sites through chat commands. You don't gotta click around dashboards or jump between tabs; you just tell your agent what you need, and it runs the checks for you.

Getting Your Lay of the Land

You want to know which sites are even connected? Use list_sites to pull a clean list of every individual website Patchstack's tracking. If you wanna see the big picture—a consolidated security summary across your entire network—run get_software_overview. That gives you a quick breakdown of all software and component versions running across every site you manage.

Deep Dive Site Audits

Need to check just one spot? You can run get_site_software against any single domain. This pulls a detailed inventory, listing every piece of software installed and what version it's running. To understand the full scope of known risks, you have two options: first, you can use search_vulnerabilities. This runs a broad search across Patchstack’s massive database using keywords or component names to find general flaws.

If you know the specific plugin or theme name, run get_component_vulnerabilities instead; this pulls every known security flaw tied directly to that piece of software. For the nitty-gritty details on a single issue, use get_vulnerability_details. You just give it a vulnerability ID or component flaw name, and it hands you the technical write-up you need.

Keeping Watch for Problems

Your agent keeps tabs on what's new. Use get_latest_vulnerabilities to pull a list of the newest flaws that just got added to Patchstack’s database—it’s good to know what risks are populating right now. For real-time threats, run get_latest_alerts. This gathers all the most recent security alerts and shows you any active firewall rules that were triggered on your sites.

You can also check the automatic updates by running get_autoupdate_settings; this retrieves the current settings that control whether vulnerable components get automatically patched.

The Workflow in Practice

Instead of manually logging into a panel to see if 'Contact Form 7' is exposed, you just ask your agent. It runs get_component_vulnerabilities for it and spits out the whole risk profile. If you suspect an old version of WordPress itself is risky, you run get_site_software, then cross-reference that version number with a search using search_vulnerabilities.

You don't have to manually check every site; your agent handles the entire inventory flow for you.

You’ll use these tools together. First, you might start by running list_sites so you know what's connected. Next, you run get_software_overview to see if anything looks wrong across the board. If that report flags a component, say 'WooCommerce', you immediately follow up with get_component_vulnerabilities for WooCommerce. Then, you check if anyone’s been poking around by running get_latest_alerts.

You can even pull the details on why that alert fired using get_vulnerability_details, linking directly back to a flaw found via search_vulnerabilities. It keeps everything in one chat window. Your agent's job is to give you the data so fast and clean that you don't waste time clicking buttons or deciphering confusing dashboards.