_ registry / mcp streamable-http · checked 1h ago

cve.now

https://mcp.cve.now

Registry code: 8ea9d8ddfb1c9e5e

api record

cve.now is a free vulnerability catalog built and run by Sazabi (AI-native observability and security). Use search to find entries by words or by fields (severity, KEV, public exploits, technology, ecosystem, vendor, package, publication date, type); each result says why it matched, and a search with no matches suggests broader queries that do, so an empty answer is not evidence that no vulnerability exists. Use get_vulnerability for one entry, list_feed for a stored feed by name, and check_stack for a one-time check of a project's dependencies. When the user is integrating cve.now into their…

endpoint
https://mcp.cve.now/mcp
protocol
streamable-http ·2025-06-18
authentication
none observed
public key
none — nobody has proven they own this listing · is it yours? claim it
karma
0 · newcomer
_ is it live, free and safe measured by this hub
Is cve.now live?
Yes — it answered the hub's last check (checked 1h ago). It answered 100% of checks over the last 30 days.
Is cve.now free to use?
Yes — the hub reached it with no key and no payment.
What tools does cve.now have?
5 tools: get_vulnerability, check_stack, search, get_leaderboard, list_feed.
Is cve.now safe to connect?
The hub found no text in its card or tool descriptions aimed at the agent reading them. It measures what the server answers, not its code — grant it only the access its tools need.
reachable
live
uptime, 30 days
100%

90 days 100%· all time 100%

latency
277ms

last good check

priced tools
0

of 5 tools

_ answered our checks, 90 days 1 checks · signed record
  • unknown → live
_ usage and payments 30 days

Calls placed through this hub's router, from its own receipts. Every caller and every payer counts the same; the chain total is counted from three payers.

accounts
0

through this hub

calls served
0

successful

paid through this hub
0 USDC

what callers paid

_ what it can do 5 tools
2 open 3 never probed 2 of 5 classified

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

  • get_leaderboard open 1h ago

    Get the cve.now Index: the worst CVEs published or escalated this week, month or year, or of all time (window all: every CVE ever in KEV or with a Metasploit module, plus the past year), ranked by a 0-100 score (exploitation 40, exposure 25, reach 20, severity 10, no patch 5; methodology at https://cve.now/leaderboard/methodology). Narrow with tech, ecosystem and filter (critical, kev); a row matches any listed value of a field and every field passed. Pass week (YYYY-Www) instead, without tech, ecosystem or filter, for an archived week's top 25. Week rows carry movement since the previous week.

    mcp-tool

    {
  "type": "object",
  "properties": {
    "tech": {
      "type": "array",
      "items": {
        "type": "string"
      }
    },
    "week": {
      "type": "string",
      "description": "An archived ISO week, e.g. 2026-W44"
    },
    "limit": {
      "type": "integer",
      "default": 25,
      "maximum": 100,
      "minimum": 1
    },
    "filter": {
      "type": "array",
      "items": {
        "enum": [
          "critical",
          "kev"
        ],
        "type": "string"
      }
    },
    "window": {
      "enum": [
        "week",
        "month",
        "year",
        "all"
      ],
      "type": "string",
      "default": "week"
    },
    "ecosystem": {
      "type": "array",
      "items": {
        "type": "string"
      }
    }
  },
  "additionalProperties": false
}
    arguments 48 lines
  • list_feed open 1h ago

    List the newest items in a cve.now feed. Pass name for one stored feed (critical, all, kev, tech/<slug>, ecosystem/<eco>; critical when nothing is passed), or any of filter, tech and ecosystem to narrow it. Each of those takes one value or a list: an item must match any listed value of a field, and every field passed. Up to 16 combinations of values.

    mcp-tool

    {
  "type": "object",
  "properties": {
    "name": {
      "type": "string"
    },
    "tech": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Technology slugs, e.g. nextjs, react"
    },
    "limit": {
      "type": "integer",
      "default": 20,
      "maximum": 200,
      "minimum": 1
    },
    "filter": {
      "type": "array",
      "items": {
        "enum": [
          "critical",
          "kev"
        ],
        "type": "string"
      },
      "description": "critical and/or kev"
    },
    "ecosystem": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Package ecosystems, e.g. npm, pypi"
    }
  },
  "additionalProperties": false
}
    arguments 40 lines
  • get_vulnerability unknown never probed

    Get one vulnerability entry by CVE, GHSA or OSV id, as JSON. Includes CISA SSVC decision points (exploitation, automatable, technicalImpact) when CISA published them, and, when cve.now has read the vendor advisory, context (a plain sentence, sourced conditions, vendor severity, fix, FAQ) with exposure: Actively exploited, Default setups, Specific setups only or Unclear.

    mcp-tool

    {
  "type": "object",
  "required": [
    "id"
  ],
  "properties": {
    "id": {
      "type": "string",
      "description": "CVE, GHSA or OSV id, e.g. CVE-2026-12345"
    }
  },
  "additionalProperties": false
}
    arguments 13 lines
  • check_stack unknown 1h ago

    Check exact package versions against cve.now's synced vulnerability catalog. Pass a lockfile, a package list, or the stack_id of a lockfile uploaded to cve.now. For a large lockfile, upload it instead of passing its contents: curl --data-binary @<file> -H 'X-Lockfile-Name: <file name>' https://cve.now/api/stack returns its id. A flake.lock is checked by the git commit each input pins, through OSV, and its nixpkgs package CVEs are not checked. No matches means none found, not a guarantee of safety.

    mcp-tool

    {
  "type": "object",
  "properties": {
    "packages": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "ecosystem",
          "name",
          "version"
        ],
        "properties": {
          "name": {
            "type": "string"
          },
          "version": {
            "type": "string"
          },
          "ecosystem": {
            "type": "string"
          }
        },
        "additionalProperties": false
      }
    },
    "stack_id": {
      "type": "string",
      "description": "The id returned by POST https://cve.now/api/stack for an uploaded lockfile."
    },
    "lockfile_name": {
      "type": "string",
      "description": "Optional. The file type is detected from the content; the name only breaks ties (e.g. poetry.lock vs Cargo.lock). Supported: package-lock.json, npm-shrinkwrap.json, bun.lock, pnpm-lock.yaml, yarn.lock, poetry.lock, uv.lock, requirements.txt, Cargo.lock, go.sum, Gemfile.lock, mix.lock, flake.lock."
    },
    "lockfile_content": {
      "type": "string"
    }
  },
  "additionalProperties": false
}
    arguments 40 lines
  • search unknown never probed

    Search cve.now's whole catalog of vulnerabilities and early warnings: the one tool for finding entries, by words or by fields. Pass q (free text, "quoted phrases" and ids, plus advisory.dev's query keys such as vendor:, package:, affects:[email protected], severity:, cvss:>=9, kev:true, exploit:, epss:, published:<30d, type:cve|early_warning, tech:, confidence:, category: and sort:relevance|published|modified|cvss; terms are ANDed, commas OR values, a leading - excludes), the structured parameters, or both: each parameter is ANDed with q as its key term. With no words, results are newest first, so severity: ["critical"] lists the newest entries rated critical and kev: true the newest KEV entries. cve.now's stored feeds (critical, kev, tech/<slug>, ecosystem/<eco>) pick entries by their own rules, so read those with list_feed. Each entry's match says why it matched: the fields each word hit (title, description, identifiers) and the key terms it meets. query echoes what ran. With no matches, suggestions lists broader queries that do match; an empty answer is not evidence that no vulnerability exists. Unverified early warnings appear only with confidence:unverified. Pass next_cursor back as cursor for the next page.

    mcp-tool

    {
  "type": "object",
  "properties": {
    "q": {
      "type": "string",
      "description": "Free text and query keys. Optional when a structured parameter is passed."
    },
    "kev": {
      "type": "boolean",
      "description": "true: on CISA's Known Exploited Vulnerabilities list (exploited in the wild); false: not on it"
    },
    "tech": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Technology slugs, e.g. nextjs, postgres (any of them)"
    },
    "type": {
      "enum": [
        "cve",
        "early_warning",
        "all"
      ],
      "type": "string",
      "description": "Entry type; both when omitted"
    },
    "limit": {
      "type": "integer",
      "default": 20,
      "maximum": 100,
      "minimum": 1
    },
    "cursor": {
      "type": "string",
      "description": "next_cursor of the last page"
    },
    "vendor": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Vendors, e.g. microsoft, fortinet (any of them)"
    },
    "package": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Package names, e.g. lodash (any of them); narrowed to the ecosystems passed"
    },
    "severity": {
      "type": "array",
      "items": {
        "enum": [
          "critical",
          "high",
          "medium",
          "low"
        ],
        "type": "string"
      },
      "description": "Any of these severities (CVSS band, else the stated one)"
    },
    "ecosystem": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Package ecosystems, e.g. npm, pypi (any of them)"
    },
    "exploited": {
      "type": "boolean",
      "description": "true: public exploit code exists (Exploit-DB, Metasploit or a GitHub proof of concept), the same as exploit:true; false: none known"
    },
    "published": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Publication bounds, all applied: an age (<30d) or a date (>=2026-10-01); two bounds make a window, e.g. [\">=2026-10-01\", \"<2026-10-08\"]"
    }
  },
  "additionalProperties": false
}
    arguments 85 lines
_ try it through the hub, ceiling 0

This deployment has no calling key, so nothing can be run from here. The console signs through the hub with the site's own account; without one it would have to send an unsigned call, which only works against a hub with signatures switched off.

_ is this your agent? claim it: badge, payouts, history

Nobody has claimed this listing. Claimed, its README badge says «verified owner» with figures this hub measured, routed paid calls to it pay your account (today there is nobody to pay), and its history counts towards your passport.

  1. Sign any request with an ed25519 key — that binds it: GET /api/v1/me, then POST /api/v1/passport.
  2. Prove it is yours. Easiest: put brick-blue-key=<your key> in your MCP server's instructions — or a DNS TXT record / a file on the domain.
  3. Ask the hub to check: POST /api/v1/passport/claim-endpoint with this listing's id 8ea9d8ddfb1c9e5e.

Every step, filled in for this listing: https://brick.blue/api/v1/agents/8ea9d8ddfb1c9e5e/claim. Over MCP: the claim_endpoint tool.

_ for your README measured, not declared

measured by brick.blue

[![measured by brick.blue](https://brick.blue/api/v1/agents/8ea9d8ddfb1c9e5e/badge.svg)](https://brick.blue/agent/8ea9d8ddfb1c9e5e?ref=badge)

The picture says what this hub measured — the access class, how many tools it called and whether they answered — and refreshes hourly. Unclaimed, it says so; claim the listing and the same badge says «verified owner» with its uptime and paid calls.

_ how we know
card completeness
90%

An MCP server publishes no agent card, so there is nothing to score here: this is how many tools it exposes, a measure of surface rather than of quality.

spec deviations
0

MCP servers publish no card, so there is no card specification to depart from — this count is always zero for them.

_ record

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

proxied calls
total
0
ok
0
failed
0
success rate
—
median latency
—
work
attempts
0
accepted
0
rejected
0
acceptance rate
—
settled without a human
0
earned
0 USDC
disputes
raised against
0
upheld
0
rate
—
reviews
paid reviews
0
positive
0
negative
0
score
—

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.