Search patents using Google BigQuery (fast, cloud-based, 100M+ patents). This is the RECOMMENDED method for patent prior art search. No local indexing required. Searches across 100M+ worldwide patents or 12M+ US patents with full text. Args: query: Search keywords or phrase limit: Maximum results...
AI agents call search_patents_bigquery to retrieve information from Claude Patent Creator without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries patent data from BigQuery with filtering capabilities. It has no side effects—it only searches and returns results. The action is purely informational (Read category). Severity is low because querying a public patent database poses minimal risk even if misused by an AI agent; the worst outcome would be information retrieval or excessive API calls.
From the tool's definition Tool description states it 'Search patents using Google BigQuery' with parameters for query, limit, country, and year filters. Returns 'List of matching' patents. No modification, deletion, execution of code, or financial operations are described.
Documented attack patterns abuse exactly the kind of access search_patents_bigquery gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Patent Creator, and nothing reaches the server without passing your rules. This is the rule we recommend for search_patents_bigquery:
{
"version": "1",
"default": "deny",
"tools": {
"search_patents_bigquery": {}
}
}search_patents_bigquery 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 patents using Google BigQuery (fast, cloud-based, 100M+ patents). This is the RECOMMENDED method for patent prior art search. No local indexing required. Searches across 100M+ worldwide patents or 12M+ US patents with full text. Args: query: Search keywords or phrase limit: Maximum results to return (default 20, max 100) country: Country code (US, EP, JP, CN, etc.) start_year: Filter by filing year >= start_year end_year: Filter by filing year <= end_year Returns: List of matching patents with title, abstract, dates, etc. It is categorised as a Read tool in the Claude Patent Creator MCP Server, which means it retrieves data without modifying state.
Register the Claude Patent Creator MCP server in PolicyLayer and add a rule for search_patents_bigquery: 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 Claude Patent Creator. Nothing to install.
search_patents_bigquery 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_patents_bigquery 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_patents_bigquery. 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_patents_bigquery is provided by the Claude Patent Creator MCP server (robthepcguy/claude-patent-creator). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 30 Claude Patent Creator tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
30 Claude Patent Creator tools catalogued and risk-classified — across an index of 42,500+ MCP servers.