Skip to content

cve-intelligence MCP server

com.cve-security/cve-intelligencecve-security.comstreamable-httpv1.7.0Scanned

CVE intelligence: exploitation (KEV/EPSS), detection coverage, fixed versions, CPE configurations.

9 tools: 9 only read.

Tools: 0 delete, 0 irreversible, 0 write, 9 read.
  • Reads private data · 4
  • Reads text others write · 0
  • Can send data out · 0
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

Decide which of this server's tools your agent can use.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the cve-intelligence MCP server can do

The cve-intelligence MCP server is published by cve-security.com, a domain the MCP Registry verified. It has 9 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.

4 tools read private data (get_cve, get_updates, get_scoreboard and 1 more).

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.

Last scanned Nov 13, 2026, version 1.7.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

9 tools

Only reads 9Show 9 read-only tools
  • get_chainsKnown Chained Vulnerabilities™: pairs of CVEs that a cited source reports were used together in one exploit chain (VulnCheck KEV entry text, Metasploit modules, SigmaHQ rules, pres…No findings

    Why it is in this group

    • The name "get_chains" says "get": a word for reading.
    Description, as the server wrote it
    Known Chained Vulnerabilities™: pairs of CVEs that a cited source reports were used together in one exploit chain (VulnCheck KEV entry text, Metasploit modules, SigmaHQ rules, press, research or academic sentences, community text judged by a local model). Each row carries both CVEs with their CISA KEV status, the claim kind (observed: the source reports attacks; potential: the source reports they can be chained), the quoted evidence with its source, URL and date, and community discussion counts, which show discussion and are not chain claims. The per-CVE record carries chains.known and chains.candidates (KCV Watch: possible chains for teams to research, pairs whose extracted exploit capabilities connect or whose records tie them together, derived and never confirmed, each with its tier, lane, reasons, basis, product, bridge, grade, a caption and the entry step where one is extracted); search_cves accepts chained=1 and chainability=1. Filters: source (vulncheck_kev, metasploit, sigma, press, research, community), since (YYYY-MM-DD, first seen), claim (observed|potential), limit (1..50).

    Annotations the server declares

    • title: Get known chains
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • get_cveFull intelligence record for one CVE: per-scorer CVSS, EPSS, CISA KEV/ransomware/SSVC, four remote-detection modalities (the checks that work over the network) plus a host-check ti…Reads private dataNo findings

    Why it is in this group

    • The name "get_cve" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "table": it reads databases.
    Description, as the server wrote it
    Full intelligence record for one CVE: per-scorer CVSS, EPSS, CISA KEV/ransomware/SSVC, four remote-detection modalities (the checks that work over the network) plus a host-check tier (self-contained Nuclei templates and Metasploit local modules that run on the system itself) and the Sigma log-detection layer, both kept out of scannable coverage, per-product fixed versions (fixed = first patched build; affected_through = the last vulnerable build, so upgrade past it), news/community coverage, intelligence summary, and bod_26_04: the BOD 26-04 Table 1 read on both exposure branches (CISA's row numbers and timelines from KEV status and CISA's SSVC Automatable and Technical impact) with CISA's KEV due date, forensic triage flag and the KEV entry's action text (kev_required_action, CISA's requiredAction field); the agency's exposure tag decides the row. other_databases lists the CVE's records in ENISA's EUVD, BDU (FSTEC of Russia), CNVD (CNCERT/CC) and JVN iPedia (JPCERT/CC and IPA) with the address of each. products lists the vendor and product pairs the page lists and the CPE entries NVD's configurations name: with cpes_source nvd_configurations, configurations carries them as NVD writes them (each entry's vulnerable flag, AND and OR operators, negate and the version bounds verbatim, up to 100 matches here, configurations_cap for more), and cpes their names; with nvd_cpe_list, cpes holds the first 25 names in sort order our NVD ingest stores, flagged by cpes_may_be_incomplete. No key required over MCP; an API key on the HTTP request (Authorization: Bearer cvs_live_…) is honored for attribution. Absence semantics: a null field means this dataset holds no such record. The source may still hold one.

    Annotations the server declares

    • title: Get CVE record
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • get_epss_moversCVEs whose EPSS exploitation probability rose the most recently.No findings

    Why it is in this group

    • The name "get_epss_movers" says "get": a word for reading.
    Description, as the server wrote it
    CVEs whose EPSS exploitation probability rose the most recently. window is "7d" (default) or "30d". Each rise is measured between same-EPSS-model-version scores, so a model release (which shifts the whole distribution) never appears as a mover. A rise raises the priority of a CVE; observed exploitation is recorded through CISA KEV. Returns cve_id, current score, the delta, KEV status and url, largest rise first.

    Annotations the server declares

    • title: Get EPSS movers
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • get_scoreboardThe Defender Scoreboard report (CC BY 4.0): exploited vs detectable vs patchable, every figure with its method, caveat and denominator, plus the corpus block and any method-change …Reads private dataNo findings

    Why it is in this group

    • The name "get_scoreboard" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "note": it reads notes.
    Description, as the server wrote it
    The Defender Scoreboard report (CC BY 4.0): exploited vs detectable vs patchable, every figure with its method, caveat and denominator, plus the corpus block and any method-change notes. Cite as "CVE Security Defender Scoreboard, cve-security.com/scoreboard".

    Annotations the server declares

    • title: Get Defender Scoreboard
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • get_sightingsField sightings: CVEs a named sensor network recorded in the last 7 or 30 days, most sighting days first.No findings

    Why it is in this group

    • The name "get_sightings" says "get": a word for reading.
    Description, as the server wrote it
    Field sightings: CVEs a named sensor network recorded in the last 7 or 30 days, most sighting days first. A field sighting is a day on which Shadowserver honeypots (cited by VulnCheck KEV and published as daily lists by CIRCL Vulnerability-Lookup) or VulnCheck canary sensors recorded traffic aimed at the CVE. Each row carries first and last sighting day, days sighted in the last 7 and 30, the sensors, and per-sensor detail including a 30-day presence strip. Presence per day, without volume; a sighting stays apart from the exploitation claims and from CISA KEV. Filters: window (7|30, default 7), kev (0|1), limit (1..50).

    Annotations the server declares

    • title: Get field sightings
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • get_updatesThe publication change stream: what this site published, stamped with OUR publish time (first_published, kev_added, detection_added, remediation_added, first_sighted, chain_added, …Reads private dataNo findings

    Why it is in this group

    • The name "get_updates" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "table": it reads databases.
    Description, as the server wrote it
    The publication change stream: what this site published, stamped with OUR publish time (first_published, kev_added, detection_added, remediation_added, first_sighted, chain_added, and the BOD 26-04 Table 1 input changes ssvc_changed, kev_due_changed, kev_triage_flag_changed, kev_notes_changed, which fire on value changes between snapshots; a timestamp refresh alone fires none). Pass since (YYYY-MM-DD, strictly-after) on the first call, then the returned next_cursor to continue. Over MCP the stream covers the last 30 days, up to 50 events per call: since defaults to 30 days ago, and an earlier since or cursor returns an error. Optional cve scopes the stream to one CVE's change history, with no 30-day cutoff; follow next_cursor past the first 50 events. Events for withdrawn CVE ids are omitted. The keyed GET /api/updates walks the full history.

    Annotations the server declares

    • title: Get update stream
    • readOnlyHint: true
    • idempotentHint: false
    • openWorldHint: false
  • query_packageCVEs affecting one open-source package, by purl (pkg:npm/lodash) or ecosystem + name (Maven names are group:artifact).No findings

    Why it is in this group

    • The name "query_package" says "query": a word for reading.
    Description, as the server wrote it
    CVEs affecting one open-source package, by purl (pkg:npm/lodash) or ecosystem + name (Maven names are group:artifact). Returns the CVE list KEV-first with each OSV version range VERBATIM: `events` plus one render-safe projection: `fixed` (the upgrade targets) or `affected_through` (the last VULNERABLE version, so upgrade past it). This tool does not evaluate version membership; compare versions on your side with your ecosystem’s own semantics. Covers CVE-linked, GitHub-reviewed OSS advisories via OSV.dev; absence is not evidence of safety.

    Annotations the server declares

    • title: Query package
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • search_cvesSearch the catalog.No findings

    Why it is in this group

    • The name "search_cves" says "search": a word for reading.
    Description, as the server wrote it
    Search the catalog. Free text (q) and/or structured filters: vendor (slug), cwe (CWE-nnn), technique (ATT&CK id, such as T1190), year ("2024,2025"), sev ("critical,high"), kev (0|1), kev_from / kev_to (ISO days, half-open CISA listing window; imply kev=1), kev_vendor (the CISA vendorProject string verbatim, such as "Microsoft"), ransomware (0|1), detect (0|1, a detection signal we track), fix (0|1; fix=0 means the fix status was computed and this dataset holds no actionable vendor fix), automatable (0|1, CISA SSVC Automatable; 1=yes, 0=CISA assessed no, unassessed CVEs match neither), sighted (7|30: a named sensor network recorded the CVE in the last 7 or 30 days, a field sighting; presence per day, apart from the exploitation claims), malware (0|1: a published source ties a named malware family, tool, campaign or ransomware group to the CVE), watch (0|1: on KEV Watch at tier 1 or 2, reported exploited by trackers other than CISA and outside CISA KEV), epss_gte (0..1), ti (total|partial: CISA SSVC Technical impact, for CVEs with a CISA assessment held), triage_flag (0|1: CISA's forensic triage flag on the KEV entry), ssvc (0|1; ssvc=0 selects rows with no CISA SSVC assessment held), bod (3df|3d|14d|60d|fsu: the BOD 26-04 Table 1 read at the stated exposure, with exposed 0|1, default 1; a mapping at that exposure, and an agency's timeline still needs its own enumeration date), eco (OSS ecosystem, such as npm or PyPI), pkg (pkg_key, such as npm/lodash; for ranges use query_package), page (1..5), limit (1..50). Filter-only queries return the /browse slice ordered KEV-first then EPSS.

    Annotations the server declares

    • title: Search CVEs
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • table1_readBOD 26-04 Table 1 read for up to 50 CVEs at a stated asset exposure.Reads private dataNo findings

    Why it is in this group

    • The name "table1_read" says "read": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "table": it reads databases.
    Description, as the server wrote it
    BOD 26-04 Table 1 read for up to 50 CVEs at a stated asset exposure. Per CVE this dataset supplies CISA KEV status and due date, CISA's SSVC Automatable and Technical impact (Vulnrichment) and CISA's forensic triage flag on the KEV entry; you supply exposure for the asset (yes, no, or unknown, which returns both branches). Each item carries the Table 1 row and timeline for the stated exposure, both branches, kev_feed (which branch reproduces CISA's due date and flag pair, if any) and the fix and detection state held, lanes listed separately. A row is a mapping; an agency's timeline for an asset also needs the agency's own enumeration date, so no date is returned beyond CISA's KEV due date. basis says where the values came from: cisa_ssvc, cisa_interim (a CVE outside KEV with no CISA assessment held takes CISA's interim values) or kev_unassessed (a KEV entry with no values held gets no rows).

    Annotations the server declares

    • title: BOD 26-04 Table 1 read
    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged untrusted-input or outbound.

Reads private data4
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:36 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
0
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools4 changed.~4 changedView diff

Install

Pinned to version 1.7.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http cve-intelligence https://cve-security.com/api/mcp
claude mcp add --transport http cve-intelligence https://cve-security.com/api/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__cve-intelligence__get_cve",
      "mcp__cve-intelligence__search_cves",
      "mcp__cve-intelligence__query_package",
      "mcp__cve-intelligence__get_updates",
      "mcp__cve-intelligence__get_scoreboard",
      "mcp__cve-intelligence__get_sightings",
      "mcp__cve-intelligence__get_chains",
      "mcp__cve-intelligence__get_epss_movers",
      "mcp__cve-intelligence__table1_read"
    ],
    "ask": [],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as cve-intelligence. Review it before you use it.

Capability tags behind 4 rules
  • mcp__cve-intelligence__get_cve private-data
  • mcp__cve-intelligence__get_updates private-data
  • mcp__cve-intelligence__get_scoreboard private-data
  • mcp__cve-intelligence__table1_read private-data

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.cve-security/cve-intelligence/verify?hash=f4a454d9a72570aa1a9946083f988f8d7c796b7971c243444f838f283c71ed2b" | jq -e '.changed == false'

Pinned hash f4a454d9a725, from Nov 13, 2026, 00:17 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.cve-security/cve-intelligence/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.cve-security/cve-intelligence
[![Anlyon Radar](https://radar.anlyon.com/badge/com.cve-security/cve-intelligence.svg)](https://radar.anlyon.com/servers/com.cve-security/cve-intelligence)

Source: MCP Registry. Synced Oct 21, 2026, 17:53 UTC.

cve-intelligence MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the cve-intelligence MCP server safe?

No scan can promise that. What its definitions show: 9 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.

Can the cve-intelligence MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the cve-intelligence MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the cve-intelligence MCP server have?

9 tools: 9 read. get_cve, search_cves, query_package, get_updates, get_scoreboard, get_sightings, get_chains, get_epss_movers and 1 more.

Does the cve-intelligence MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the cve-intelligence MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the cve-intelligence MCP server changed its tools?

Yes, once since Anlyon Radar started tracking it, most recently on Oct 31, 2026. Every change is on this page and in its Atom feed.