mcpField
C
Circl Vulnerability Lookup

Circl Vulnerability Lookup

Remote
pipeworx.ioSecurity0Updated 5d ago

About Circl Vulnerability Lookup

Circl Vulnerability Lookup is an MCP server for querying vulnerability records aggregated by CIRCL from multiple feeds. It resolves IDs across formats and provides normalized summaries alongside optional raw records.

Key features

  1. 01Look up vulnerabilities by ID
  2. 02Search by CPE vendor and product
  3. 03Retrieve recently updated records
  4. 04Normalize summaries across record schemas
  5. 05Include original upstream records

Use cases

  1. 01Check a CVE or GHSA identifier
  2. 02Review vulnerabilities for a CPE product
  3. 03Monitor recently published advisories

MCP support

Exposes vulnerability lookup, search, and recent-record tools to MCP clients. It is available through the Pipeworx gateway or as a local stdio MCP server.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://gateway.pipeworx.io/circl-vulnerability-lookup/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

@pipeworx/circl-vulnerability-lookup

Vulnerability records from CIRCL's Vulnerability-Lookup service, which resolves one id against several upstream feeds at once — MITRE CVE, NVD, GitHub Security Advisories, PySec and vendor CSAF advisories.

Part of Pipeworx — an MCP gateway connecting AI agents to 1875+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

  • circl_vuln(id, include_raw?) — one vulnerability by id. Accepts CVE-, GHSA-, PYSEC- and vendor advisory ids alike. Returns a normalised summary (title, description, CVSS score/vector, affected products, references, dates) plus the untouched upstream record.
  • circl_vuln_search(vendor, product, limit?) — every vulnerability recorded against a CPE vendor/product pair, newest first, tagged with the feed each record came from.
  • circl_vuln_recent(limit?) — the most recently published or updated records across every aggregated feed.

Auth

Keyless.

Data sources

Operated by CIRCL (Computer Incident Response Center Luxembourg).

Traps

  • An unknown id answers {} with HTTP 200. That is a silent zero — parsed naively it reads as a successful empty record. circl_vuln throws a named miss instead.
  • The real paths are not the ones the names suggest. Search is /api/search/{vendor}/{product} and the recent feed is /api/last/{n}. /api/vulnerability/search/... and /api/vulnerability/last are both 404.
  • Records arrive in three different schemas, and one /api/last response mixes all three:
    • CVE JSON 5.x — cveMetadata + containers.cna
    • OSV — id / details / affected / severity
    • CSAF — document / product_tree / vulnerabilities summarize() normalises the shared fields across all three; raw keeps the original. Do not assume containers.cna exists.
  • CVSS lives under a version-specific key (cvssV3_1, cvssV4_0, …) inside containers.cna.metrics[], so the code scans the metric objects rather than reaching for a fixed path. OSV records carry only a vector string in severity[].score, with no numeric base score.
  • Vendor/product are matched against CPE strings, not marketing names — use microsoft/windows_10, not Microsoft Windows 10.
  • CIRCL rate-limits by source IP, and publishes the limits at https://vulnerability.circl.lu/.well-known/api-policy.json: 20 requests per minute anonymously, 40 with an X-API-KEY. It does send Retry-After (observed: 59) plus X-RateLimit-Limit/-Remaining/-Reset, and the 429 error here quotes it. This pack is keyless and stays inside the anonymous bucket — one tool call is one upstream request. If we ever need the higher tier, it is a header, not a rebuild.
  • The 429 is per backend worker, not global. Measured 2026-09-17: three consecutive anonymous requests from one IP reported x-ratelimit-remaining of 0, then 19, then 3, and one URL kept 429-ing while the next id answered 200 instantly. So a 429 here frequently says nothing about your actual usage — circlJson() retries up to 3 times before surfacing it. Do not read a single 429 as "we are being throttled".

Related packs

nvd (NVD directly, platform-keyed) and osv (OSV.dev, open-source package advisories) cover single upstreams. This pack is the cross-feed id resolver: use it when you have an id and do not know which body published it.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "circl-vulnerability-lookup": {
      "url": "https://gateway.pipeworx.io/circl-vulnerability-lookup/mcp"
    }
  }
}
What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/circl-vulnerability-lookup/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1875+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/circl_vuln \
  -H 'Content-Type: application/json' \
  -d '{"id":"CVE-2014-0160","include_raw":false}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/circl_vuln. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

{
  "mcpServers": {
    "circl-vulnerability-lookup": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-circl-vulnerability-lookup"]
    }
  }
}

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-circl-vulnerability-lookup

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Circl Vulnerability Lookup data" })

The gateway picks the right tool and fills the arguments automatically.

More

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.