Search NVD for CVEs by keyword and optional severity filter. Args: query: Keyword to search (letters, numbers, spaces, hyphens, dots — max 200 chars) severity: Optional CVSS v3 severity filter: NONE, LOW, MEDIUM, HIGH, CRITICAL limit: Max results (1–50, default 10)
AI agents call search_cves to retrieve information from CVE MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a read-only query against the National Vulnerability Database (NVD) to retrieve CVE information based on search criteria. It retrieves and returns vulnerability intelligence data without creating, modifying, deleting, or executing any operations. The parameters (query, severity filter, limit) control only what data is returned, not any actions taken.
From the tool's definition Tool description states 'Search NVD for CVEs by keyword' with parameters for query, optional severity filter, and result limit. The function name 'search_cves' and description indicate data retrieval with no modification or side effects.
Documented attack patterns abuse exactly the kind of access search_cves gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and CVE MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for search_cves:
{
"version": "1",
"default": "deny",
"tools": {
"search_cves": {}
}
}search_cves is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Search NVD for CVEs by keyword and optional severity filter. Args: query: Keyword to search (letters, numbers, spaces, hyphens, dots — max 200 chars) severity: Optional CVSS v3 severity filter: NONE, LOW, MEDIUM, HIGH, CRITICAL limit: Max results (1–50, default 10). It is categorised as a Read tool in the CVE MCP Server MCP Server, which means it retrieves data without modifying state.
Register the CVE MCP Server MCP server in PolicyLayer and add a rule for search_cves: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches CVE MCP Server. Nothing to install.
search_cves is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the search_cves rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for search_cves. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
search_cves is provided by the CVE MCP Server MCP server (mukul975/cve-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from CVE MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
27 CVE MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.