MCP Observatory
github not analyzable

rozetyp/vuln-intel-mcp

github

Agent-native CVE intelligence as a hosted MCP server: ~332k vulnerabilities fused from NVD, CISA KEV, FIRST EPSS, OSV/GHSA and SSVC, ranked exploitation-first, with a CVE-claim fact-checker that catches the CVEs your agent makes up.

maintainer
rozetyp
license
NOASSERTION
first seen
2026-06-14
last seen
2026-06-15
releases · 30d
0
short id

Drift inferred · capture-to-capture

No drift recorded — single capability capture; advisories appear once its surface changes.

capabilities 9 tools
transport http counts 9 tools · 0 res · 0 prompts permission surface via README inference

tools

  • check_technology

    Map a product or stack to ranked CVEs, de-duped across NVD CPE and OSV.

  • corpus_stats

    Corpus size and data freshness.

  • enrich_cve

    One CVE, full dossier: CVSS, KEV, EPSS, SSVC, affected versions, live PoC.

  • find_recent_high_risk

    What is newly dangerous: KEV / high-EPSS, optionally per product.

  • find_similar_vulns

    Semantic, by-mechanism search from a concept or a seed CVE.

  • hunt_plan

    Turn a recon'd stack into a ranked dig-order with recurring bug-class loci.

  • search_public_code

    Find where an exact code string appears across public repositories.

  • search_vulns

    Exact full-text / CWE search across CVE and advisory summaries.

  • verify_cve_claim

    Fact-check a CVE assertion; catch hallucinations and wrong attributes.

code evidence — · github-tarball
evidence-backed findings quoted directly from the published source artifact — not inferred

last analysis: no-source

No code evidence — the analyzed source reached for no tracked permissions, tools, or hooks.