The Problem

Vulnerability intelligence is scattered across multiple databases. NVD has CVE details. EPSS tells you exploitation probability. CISA KEV tracks actively exploited vulns. GitHub Advisory covers open source packages. OSV maps vulnerabilities to specific package versions. No single tool aggregates them, and none work with AI agents.

Traditional workflow:
  search NVD for CVE details          →  navigate a clunky web UI
  check EPSS for exploitation risk    →  separate API, separate format
  look up CISA KEV status             →  download a JSON feed manually
  search GitHub advisories            →  yet another interface
  query OSV for package impact        →  different API, different schema
  correlate everything                →  copy-paste into a spreadsheet
  ──────────────────────────────────
  Total: 30+ minutes per CVE, longer for bulk triage

cve-mcp gives your AI agent 23 tools via the Model Context Protocol. The agent queries all five sources in parallel, correlates data, calculates risk scores, and tells you exactly what matters.

With cve-mcp:
  You: "Prioritize these 10 CVEs by actual exploitation risk"

  Agent: → fetches CVSS scores from NVD
         → gets EPSS exploitation probability for each
         → checks CISA KEV for actively exploited
         → cross-references GitHub advisories for patches
         → "3 are critical: CVE-2024-3400 (EPSS 97%, in KEV),
            CVE-2023-44487 (HTTP/2 rapid reset, EPSS 96%),
            CVE-2021-44228 (Log4Shell, EPSS 97%, in KEV).
            Here are patches and affected versions..."

How It's Different

Existing tools give you raw data. cve-mcp gives your AI agent the ability to reason about vulnerabilities.

Quick Start

Option 1: npx (no install)

npx cve-mcp

Option 2: Clone

git clone https://github.com/badchars/cve-mcp.git
cd cve-mcp
bun install

Environment variables (optional)

# Increases NVD rate limit from 5 to 50 requests per 30 seconds
export NVD_API_KEY=your-nvd-api-key

# Enables GitHub Advisory search (60 → 5000 requests/hour)
export GITHUB_TOKEN=ghp_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Both are optional. The server works without them, just with lower rate limits.

Connect to your AI agent

# With npx
claude mcp add cve-mcp -- npx cve-mcp

# With local clone
claude mcp add cve-mcp -- bun run /path/to/cve-mcp/src/index.ts

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "cve-mcp": {
      "command": "npx",
      "args": ["cve-mcp"],
      "env": {
        "NVD_API_KEY": "optional-key",
        "GITHUB_TOKEN": "optional-token"
      }
    }
  }
}

Same JSON config format. Point the command to npx cve-mcp or your installation path.

Start querying

You: "What do you know about CVE-2024-3400?"

That's it. The agent handles the rest.

What The AI Can Do

Incident Response

You: "We got hit by CVE-2024-3400. Give me everything."

Agent: → cve_enrich {cveId: "CVE-2024-3400"}
       → NVD: PAN-OS command injection, CVSS 10.0
       → EPSS: 97.2% exploitation probability
       → KEV: Added 2024-04-12, due 2024-05-01
       → GHSA: No open source advisory (commercial product)
       → exploit_search: 23 PoC repos on GitHub
       → "Critical. Actively exploited in the wild.
          PAN-OS GlobalProtect, versions < 10.2.9-h1.
          Patch immediately. CISA deadline: May 1."

Vulnerability Prioritization

You: "Prioritize these CVEs from our scan: CVE-2021-44228, CVE-2024-3400,
      CVE-2023-44487, CVE-2024-21762, CVE-2023-4966"

Agent: → cve_prioritize {cves: [...]}
       → Ranks by CVSS × EPSS × KEV multiplier
       → "#1: CVE-2024-3400 (risk: 19.44, CVSS 10.0, EPSS 97%, KEV)
          #2: CVE-2021-44228 (risk: 19.40, CVSS 10.0, EPSS 97%, KEV)
          #3: CVE-2023-4966 (risk: 19.10, CVSS 9.4, EPSS 97%, KEV)
          All 5 are in CISA KEV — patch all immediately."

Dependency Audit

You: "Check if lodash 4.17.20 and django 3.2.0 have known vulnerabilities"

Agent: → osv_batch {queries: [
          {package: "lodash", version: "4.17.20", ecosystem: "npm"},
          {package: "django", version: "3.2.0", ecosystem: "PyPI"}
       ]}
       → "lodash 4.17.20: 3 vulnerabilities (prototype pollution)
          django 3.2.0: 12 vulnerabilities (SQL injection, XSS)
          Upgrade lodash to 4.17.21+, django to 4.2+"

Threat Monitoring

You: "What are the most likely to be exploited CVEs right now?"

Agent: → cve_trending {limit: 10, minEpss: 0.9}
       → "Top 10 by exploitation probability:
          1. CVE-2024-3400 — PAN-OS (EPSS 97.2%, CVSS 10.0, KEV)
          2. CVE-2023-44487 — HTTP/2 Rapid Reset (EPSS 96.5%, CVSS 7.5, KEV)
          ..."

CVSS Deep Dive

You: "Break down this CVSS vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H"

Agent: → cvss_parse {vector: "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H"}
       → "Score: 10.0 (CRITICAL)
          Network attack vector, no privileges required, no user interaction.
          Scope changed — can impact other components.
          Full compromise: confidentiality, integrity, and availability."

Tools Reference (23 tools)

| Tool | Description | |------|-------------| | nvd_search | Search CVEs by keyword, severity, CWE, date range | | nvd_get | Get full CVE details (CVSS, CWE, CPE, references) | | nvd_recent | Recently published/modified CVEs | | cve_by_product | Search CVEs by product name (CPE keyword match) |

| Tool | Description | |------|-------------| | epss_score | EPSS exploitation probability for one or more CVEs | | epss_top | Top CVEs by exploitation probability |

| Tool | Description | |------|-------------| | kev_check | Check if CVE(s) are in CISA Known Exploited Vulnerabilities catalog | | kev_search | Search KEV by vendor, product, or keyword | | kev_recent | Recently added KEV entries |

| Tool | Description | |------|-------------| | ghsa_search | Search GitHub security advisories by keyword, ecosystem, severity | | ghsa_get | Get advisory details by GHSA ID or CVE ID |

| Tool | Description | |------|-------------| | osv_query | Query vulnerabilities for a specific package version | | osv_get | Get vulnerability details by OSV/GHSA/CVE ID | | osv_batch | Batch query multiple packages at once |

| Tool | Description | |------|-------------| | exploit_search | Search for public PoC exploits (GitHub repositories) |

| Tool | Description | |------|-------------| | cwe_lookup | Look up CWE weakness by ID or search by keyword |

| Tool | Description | |------|-------------| | cvss_parse | Parse and explain a CVSS v3.1 vector string with score calculation |

| Tool | Description | |------|-------------| | cve_enrich | Full enrichment: NVD + EPSS + KEV + GHSA + OSV in parallel | | cve_prioritize | Rank CVEs by risk (CVSS × EPSS × KEV multiplier) | | cve_trending | Currently trending CVEs by exploitation probability | | cve_compare | Side-by-side comparison of two CVEs | | cve_list_sources | List all data sources and their availability | | cve_report | Generate a markdown vulnerability report |

Data Sources

| Source | Auth | What it provides | |--------|------|-----------------| | NVD | Optional NVD_API_KEY | CVE details, CVSS scores, CWE mappings, CPE affected products, references | | EPSS | None | Exploitation probability score (0-1) and percentile ranking | | CISA KEV | None | Known exploited vulnerabilities with remediation deadlines | | GitHub Advisory | Optional GITHUB_TOKEN | Open source security advisories, affected packages, severity | | OSV | None | Package-level vulnerability data across 16+ ecosystems |

Risk Score Formula

Risk Score = CVSS Base Score × EPSS Score × KEV Multiplier

Where:
  CVSS Base Score  = 0-10 (severity from NVD)
  EPSS Score       = 0-1 (exploitation probability from FIRST)
  KEV Multiplier   = 2 if in CISA KEV, 1 otherwise

This gives a practical risk score that balances severity (CVSS) with real-world exploitation likelihood (EPSS) and known active exploitation (KEV).

Architecture

src/
├── index.ts                    Entry point + MCP stdio
├── types/
│   └── index.ts                ToolDef, ToolContext, ToolResult, API types
├── protocol/
│   ├── tools.ts                23 tool definitions (Zod schemas)
│   └── mcp-server.ts           MCP server + stdio transport
├── nvd/
│   ├── index.ts                NVD API v2 — search, get, recent
│   └── cpe.ts                  Product/CPE search
├── epss/
│   └── index.ts                EPSS — score, top
├── kev/
│   └── index.ts                KEV — check, search, recent (cached)
├── ghsa/
│   └── index.ts                GitHub Advisory — search, get
├── osv/
│   └── index.ts                OSV — query, get, batch
├── exploit/
│   └── index.ts                PoC search via GitHub repos
├── cwe/
│   └── index.ts                CWE database (40+ entries embedded)
├── cvss/
│   └── index.ts                CVSS v3.1 parser + score calculator
├── meta/
│   ├── enrich.ts               Full CVE enrichment (all sources)
│   ├── prioritize.ts           Risk-based CVE ranking
│   ├── trending.ts             Trending CVEs by EPSS
│   ├── compare.ts              Side-by-side CVE comparison
│   └── sources.ts              Data source health check
└── utils/
    ├── rate-limiter.ts          Queue-based rate limiter (NVD)
    └── cache.ts                 TTL cache (NVD results)

Design decisions:

• Intelligence, not audit — Unlike cloud-audit-mcp and github-security-mcp, this is a data tool. No CheckResult, no findings accumulation. Each query is independent and stateless.
• Parallel enrichment — cve_enrich calls all 5 sources via Promise.allSettled. If one source is down, the rest still return data.
• Shared rate limiter — All NVD modules share a single RateLimiter instance (6s between requests) to avoid 429 errors.
• KEV caching — The full KEV catalog (~1200 entries, ~200KB) is loaded once and cached in memory with 1-hour TTL.
• NVD result caching — Individual CVE lookups are cached with 10-minute TTL to avoid redundant API calls during enrichment.
• Embedded CWE database — 40+ top CWEs from OWASP/MITRE embedded as static data. No external API needed.
• Full CVSS calculator — CVSS v3.1 base score calculation algorithm implemented locally. Parses any vector string and explains each metric.
• 2 dependencies — @modelcontextprotocol/sdk and zod. Nothing else.

Limitations

• NVD API without NVD_API_KEY is limited to 5 requests per 30 seconds. Set the key for production use
• GitHub Advisory search without GITHUB_TOKEN is limited to 60 requests per hour
• Exploit search uses GitHub repository search which has its own rate limits
• CVSS parser supports v3.1 only (v2 and v4 vectors are not parsed, though v2 scores are returned from NVD)
• CWE database is embedded (40+ entries) rather than comprehensive (1000+ in full MITRE database)
• macOS / Linux (Windows not tested)

Part of the MCP Security Suite

| Project | Domain | Tools | |---|---|---| | hackbrowser-mcp | Browser-based security testing | 39 tools, Firefox, injection testing | | cloud-audit-mcp | Cloud security (AWS/Azure/GCP) | 38 tools, 60+ checks | | github-security-mcp | GitHub security posture | 39 tools, 45 checks | | cve-mcp | Vulnerability intelligence | 23 tools, 5 sources |