Vulners MCP: CVE triage and package audits from your AI app
Look up a CVE, audit a software version or check a Linux server's packages against the Vulners vulnerability database, all from your AI app. Vulners' MCP server wraps the Vulners.com API in seven tools. ContextGate connects it to Claude, ChatGPT and Cursor, and policies check each call, so a hostname or key caught in a pasted package list can be blocked before it leaves.
- Ready to connect
- Free
- 7 example tools
How to add Vulners to Claude, ChatGPT or Cursor
- 1Click Connect Vulners free. There's no sign-in, so its tools are ready at once.
- 2All seven tools look things up. The two audit tools, `audit_software` and `audit_linux_packages`, send your software versions to Vulners; switch them off if your package lists shouldn't leave your network, and keep `bulletin_by_id` for public CVE lookups.
- 3Add ContextGate to your AI app as a connector or MCP server. The steps for each app are on the Works with page.
- 4Ask your AI app for the CVSS and EPSS scores of a recent CVE, then open Activity to see the `bulletin_by_id` call.
Vulnerability research with Vulners from Claude or Cursor
- Fetch a bulletin by CVE or Vulners ID with `bulletin_by_id` and get a short triage card: title, CVSS3, EPSS and related CVEs.
- Search the Vulners knowledge base with Lucene queries through `search_lucene` when you don't have an ID yet.
- Check exact software versions for known vulnerabilities with `audit_software`, or audit RPM and DEB packages for a distro and version with `audit_linux_packages`.
- Find the CPE string for a vendor and product with `search_cpe`.
What leaves your network, and what's recorded
- Switch off the audit tools to keep your AI app to public CVE data; switched-off tools are hidden and refused if called.
- A policy can block secrets or internal hostnames in a package list before an audit call goes out.
- Activity records every Vulners lookup, so you can see which CVEs your team checked and when.
What you get with Vulners and ContextGate
- Your AI apps can use VulnersClaude, Cursor, ChatGPT and other AI apps reach Vulners through the one ContextGate connection you add to them.
- You choose which tools they can useSwitch off any Vulners tool you don't want used. It's hidden from your AI apps and refused if called.
- Every call is checked and recordedYour policies check each call (for example, personal data is hidden and secrets are blocked), and Activity keeps a record of it.
ContextGate sees only the tool calls that go through it, not your AI app's built-in tools or your chats.
Vulners tools
Examples of what your AI apps can do in Vulners. You'll see the exact list of tools after you connect it.