EchelonGraph CVE & Exposure logo

EchelonGraph CVE & Exposure

Keyless MCP access to CVE intelligence, KEV, EPSS, SBOM checks, advisories, and Shodan-derived exposure data.

Install EchelonGraph CVE & Exposure

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "echelongraph-cve-exposure": {
      "command": "npx",
      "args": [
        "-y",
        "echelongraph-mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— the server itself is confirmed live (we called its tools/list endpoint). The stdio install command is a separate, ongoing check.

Overview

EchelonGraph CVE & Exposure MCP server connects AI clients to EchelonGraph’s public vulnerability feed and exposure radars. Its read-only tools cover CVE search and enrichment, CISA KEV updates, EPSS history, vendor advisories, package and SBOM checks, remediation data, and CWE lookups. Per-CVE and aggregate exposure results derive from Shodan data and describe sampled internet-facing services rather than exploitability or an internet-wide census. Use it for vulnerability triage, dependency assessment, and security briefings when a keyless MCP connection is useful.

Use cases

  • Triage a CVE using severity, KEV, EPSS, exploit, remediation, and exposure data
  • Check whether a package or product version is affected
  • Review an SBOM or lockfile for vulnerable dependencies
  • Prepare a weekly brief from recent CISA KEV additions

Key features

  • Searches and summarizes fused CVE intelligence
  • Checks packages, CPE products, SBOMs, and lockfiles
  • Reports CISA KEV and EPSS history
  • Retrieves vendor advisories and remediation evidence
  • Provides CWE and exploit enrichment
  • Shows Shodan-derived per-CVE and aggregate exposure

Tools16

What EchelonGraph CVE & Exposure lets your agent do. Expand a tool for its parameters.

Verified live~12.1k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked

Recent health check succeeded.

8/8 checks healthy over the last 28d

What it can do

From the 16 tools this server returned when we ran it; 16 declare their own safety hints.

Handles credentials · 1Makes network requests · 16Read-only tools · 16
Network at startup3 connections
Files written at startupNone
Install scriptsNone
npm provenanceSigned build
Package age123 days
Maintainers1

Observed 2026-10-26 in an isolated sandbox. A description of reach, not a security audit.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 16d ago via OSV.dev · echelongraph-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars21Stargazers on the source repository.
npm downloads9.4kPackage downloads in the last 30 days.
Last commitTodayMost recent push to the default branch.
Tools exposed16Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How EchelonGraph CVE & Exposure works

What EchelonGraph CVE & Exposure does

EchelonGraph CVE & Exposure combines vulnerability records from NVD, MITRE-CNA pre-NVD records, CISA KEV, EPSS, and GitHub GHSA into a queryable feed. It exposes 16 MCP tools for looking up individual CVEs, searching and summarizing the feed, reviewing recently added KEV entries, inspecting EPSS changes, and retrieving CWE classifications.

The server also handles software-impact checks. You can test a product and version through CPE data, evaluate a package through OSV ecosystem data, scan CycloneDX or SPDX SBOMs, or inspect supported lockfiles and pinned manifests. Advisory tools add vendor-published guidance, remediation records, exploit references, affected packages, fixed versions, and CISA KEV actions.

Exposure tools report aggregate counts for internet-facing services associated with tracked vulnerabilities, exposed databases, leaked credentials, shadow AI services, and MCP-server observations. The per-CVE view includes country and product breakdowns where available; the radar view provides aggregate totals.

How it works

A local installation runs as a read-only MCP process and queries EchelonGraph only when a tool call requires data. The service is free and keyless, with no API key or authentication required. A hosted MCP endpoint is also available for clients that support remote connections.

Feed and radar values include measurement or coverage metadata such as state, method, freshness, timestamps, and notes. These fields distinguish assessed results from unassessed lookups, incomplete coverage, and unavailable measurements. Results can be large; responses are structured and are cut around 30,000 characters unless a call requests the full response.

Exposure counts come from Shodan-derived observations. The radar checks tracked products periodically, retains qualifying service records for a limited period, and counts distinct IP-and-port pairs. A count is therefore a version-based inference over the radar’s sample, not an exploit test or a census of the public internet.

Setup and configuration

The local package requires Node.js 20 or later and can be started with npx:

config.json
{
  "mcpServers": {
    "echelongraph": {
      "command": "npx",
      "args": ["-y", "echelongraph-mcp"]
    }
  }
}

Add this entry to the relevant MCP configuration for Claude Desktop or Cursor. VS Code uses a servers key instead of mcpServers. Claude Code can register the same stdio command with claude mcp add --transport stdio. Claude Desktop also supports a downloadable extension that runs without a separate Node.js installation, or a custom connector using https://mcp.echelongraph.io/mcp.

The package requires no environment variables, credentials, or global installation. Restart the client after changing its configuration so it reloads the server.

Tools and capabilities

  • Search, summarize, and retrieve CVE records with severity, CVSS, EPSS, KEV, CPE, references, and remediation fields.
  • Inspect exploit sources, affected packages, fixed versions, vendor advisories, and CWE membership.
  • Review recent CISA KEV additions and historical EPSS changes.
  • Check CPE products, registry packages, SBOMs, lockfiles, and pinned manifests.
  • Query per-CVE exposure and aggregate radar counts derived from Shodan and other stated sources.
  • Use bundled workflows for CVE triage, KEV briefings, affected-version checks, SBOM reviews, and workload triage.

Limitations and notes

An unassessed or undetermined result is not a clean result. A zero can represent missing assessment or lack of recorded evidence, depending on the tool, and does not prove that a vulnerability, fix, exploit, or exposure is absent. Vendor remediation text is relayed from its source and is not tested by EchelonGraph.

SBOM and manifest checks send extracted package URLs to the API; local package documents and filenames are not sent when using the local process. Supported manifest formats are limited to the documented lockfiles and pinned formats, while range-based project files are refused. Calls have input, rate, pagination, and output-size limits, and truncated responses identify omitted data where possible.

Shodan-derived data is owned by Shodan. Exposure results may remain counted while an open port is refreshed even if the service has changed, and CVEs outside the radar’s tracked set can be marked not assessed.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about EchelonGraph CVE & Exposure

EchelonGraph CVE & Exposure is an MCP server that connects AI clients to EchelonGraph’s public vulnerability and exposure data. Its main tools search CVEs, inspect KEV and EPSS records, check packages and SBOMs, retrieve advisories and remediation, and report Shodan-derived exposure counts. It is read-only, free, and keyless.

Install it locally with `npx -y echelongraph-mcp` and add that command to your client’s MCP configuration. Node.js 20 or later is required for the local package. Claude Desktop also offers a downloadable extension and a custom connector for the hosted endpoint.

Yes, EchelonGraph CVE & Exposure works with Cursor through its MCP configuration. Add an `echelongraph` entry using `npx -y echelongraph-mcp` to `~/.cursor/mcp.json` for all projects or `.cursor/mcp.json` for one project.

Yes, it can scan CycloneDX or SPDX SBOMs and supported lockfiles or pinned manifests. It returns affected, not affected, undetermined, or not assessed results. Unsupported range-based project files are refused, and an unassessed or undetermined result is not a clean finding.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewEchelonGraph CVE & Exposure AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/echelongraph-cve-exposure?style=directory)](https://allmcps.com/mcp/echelongraph-cve-exposure)
HTML Embed
<a href="https://allmcps.com/mcp/echelongraph-cve-exposure"><img src="https://allmcps.com/api/badge/echelongraph-cve-exposure?style=directory" alt="EchelonGraph CVE & Exposure on AllMCPs" /></a>