Skip to main content
Glama

get_most_cited_patents

Identify influential patents by retrieving the most frequently cited ones within specific technology areas using keywords or IPC codes, with filtering by date and authority.

Instructions

View the top patents cited most frequently by others, indicating influential or core technology. Returns at most Top 10 patents. Note: Search must contain either keywords or IPC. If both are provided, IPC is prioritized.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
keywordsNoKeywords to search within patent title and abstract/summary. Supports AND, OR, NOT logic. Example: "mobile phone AND (screen OR battery)"
ipcNoPatent IPC classification code. Used to specify a particular technology field.
apply_start_timeNoPatent application start year (yyyy format). Filters by application filing date.
apply_end_timeNoPatent application end year (yyyy format). Filters by application filing date.
public_start_timeNoPatent publication start year (yyyy format). Filters by publication date.
public_end_timeNoPatent publication end year (yyyy format). Filters by publication date.
authorityNoPatent authority code (e.g., CN, US, EP, JP). Filters by patent office. Use OR for multiple, e.g., "US OR EP".

Implementation Reference

  • The handler function implementing the core logic for the 'get_most_cited_patents' tool. It constructs query parameters from input args and delegates to the shared PatSnap API caller for the 'most-cited' endpoint.
    async function getMostCitedPatents(args: BasePatentArgs): Promise<ServerResult> {
      const params = buildCommonSearchParams(args);
      return callPatsnapApi('most-cited', params, 'get most cited patents');
    }
  • The base input schema used by the 'get_most_cited_patents' tool (and others), defining properties like keywords, ipc, date ranges, and authority.
    const basePatentInputSchema = {
        type: 'object' as const, // Use 'as const' for stricter type checking
        properties: {
            keywords: { type: 'string', description: 'Keywords to search within patent title and abstract/summary. Supports AND, OR, NOT logic. Example: "mobile phone AND (screen OR battery)"' },
            ipc: { type: 'string', description: 'Patent IPC classification code. Used to specify a particular technology field.' },
            apply_start_time: { type: 'string', description: 'Patent application start year (yyyy format). Filters by application filing date.' },
            apply_end_time: { type: 'string', description: 'Patent application end year (yyyy format). Filters by application filing date.' },
            public_start_time: { type: 'string', description: 'Patent publication start year (yyyy format). Filters by publication date.' },
            public_end_time: { type: 'string', description: 'Patent publication end year (yyyy format). Filters by publication date.' },
            authority: { type: 'string', description: 'Patent authority code (e.g., CN, US, EP, JP). Filters by patent office. Use OR for multiple, e.g., "US OR EP".' }
        },
        // Add a note about requiring keywords or IPC for most tools
        description: "Requires either 'keywords' or 'ipc' to be specified for a meaningful search. If both are provided, IPC is prioritized by the API."
    };
  • src/index.ts:351-355 (registration)
    Registration of the tool in the ListTools response, including its name, description, and reference to the input schema.
    {
      name: 'get_most_cited_patents',
      description: 'View the top patents cited most frequently by others, indicating influential or core technology. Returns at most Top 10 patents. Note: Search must contain either keywords or IPC. If both are provided, IPC is prioritized.',
      inputSchema: basePatentInputSchema
    },
  • src/index.ts:398-398 (registration)
    Mapping of the tool name to its handler function in the internal dispatch object used by the CallToolRequest handler.
    'get_most_cited_patents': getMostCitedPatents,
  • Shared helper function that performs the actual API call to PatSnap's insights endpoint, handles authentication, errors, and formats the response. Called by the handler with endpoint 'most-cited'.
    async function callPatsnapApi(endpoint: string, params: URLSearchParams, errorContext: string): Promise<ServerResult> {
        const token = await getAccessToken(); // Will use cached token if available and valid
        const url = `${PATSNAP_API_BASE_URL}/insights/${endpoint}?${params.toString()}`;
        console.log(`Calling PatSnap API: ${url}`); // Log the request URL (consider using a proper logger)
    
        let response: Response;
        try {
            response = await fetch(url, {
                method: 'GET',
                headers: {
                    // 'Content-Type': 'application/json', // Typically not needed for GET
                    'Authorization': `Bearer ${token}`
                }
                // Consider adding a timeout
                // signal: AbortSignal.timeout(15000) // e.g., 15 seconds timeout
            });
        } catch (error) {
            console.error(`Network error calling PatSnap API endpoint ${endpoint}:`, error);
            throw new McpError(503, `Network error connecting to PatSnap API (${endpoint}): ${error instanceof Error ? error.message : String(error)}`);
        }
    
    
        if (!response.ok) {
            let errorText = `Status code ${response.status}`;
            try {
                errorText = await response.text();
            } catch (e) {
                console.error("Failed to read error response body:", e);
            }
            console.error(`API Error (${response.status}) for ${endpoint}: ${errorText}`); // Log error details
            // Invalidate cache on auth errors (401 Unauthorized, 403 Forbidden)
            if (response.status === 401 || response.status === 403) {
                 cachedToken = null;
                 console.log('Authentication error detected, clearing token cache.');
            }
            // Map common PatSnap error codes to potentially more user-friendly messages if desired
            // Example: if (errorText.includes("67200002")) { throw new McpError(429, "PatSnap API quota exceeded."); }
            throw new McpError(response.status, `Failed to ${errorContext}: ${errorText}`);
        }
    
        let json: PatsnapApiResponse; // Use interface type
        try {
            json = await response.json() as PatsnapApiResponse; // Type assertion
        } catch (error) {
            console.error(`Error parsing JSON response from ${endpoint}:`, error);
            throw new McpError(500, `Failed to parse JSON response from PatSnap API (${endpoint}): ${error instanceof Error ? error.message : String(error)}`);
        }
    
        // Basic check for PatSnap's own error structure within a 200 OK response
        if (json && typeof json.status === 'boolean' && json.status === false && json.error_code !== 0) {
            console.error(`PatSnap API returned error within successful response for ${endpoint}: Code ${json.error_code}, Msg: ${json.error_msg}`);
            // You might want to map these internal errors to McpError as well
            throw new McpError(400, `PatSnap API Error (${json.error_code || 'N/A'}): ${json.error_msg || 'Unknown error'}`);
        }
    
        return {
            content: [
                {
                    type: 'text',
                    // Return the raw JSON response as text, formatted for readability
                    text: JSON.stringify(json, null, 2)
                }
            ]
        };
    }
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. It adds useful context about the search requirement (keywords or IPC) and prioritization rule, but does not cover other behavioral aspects such as rate limits, authentication needs, or what happens with invalid inputs. The description does not contradict any annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is appropriately sized with two sentences that are front-loaded with key information (purpose and constraints). Every sentence earns its place by clarifying the tool's function and usage rules without unnecessary details.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's moderate complexity (7 parameters, no output schema, no annotations), the description is adequate but has gaps. It explains the search requirement and result limit, but does not detail the return format (e.g., what fields are included in the top patents) or error handling, which could be important for an AI agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema description coverage is 100%, providing detailed descriptions for all 7 parameters. The description adds minimal value beyond the schema by reiterating the requirement for 'keywords or IPC' and IPC prioritization, which is already covered in the schema's description. This meets the baseline for high schema coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose with specific verbs ('view', 'returns') and resources ('top patents cited most frequently'), distinguishing it from siblings like 'get_most_litigated_patents' or 'get_patent_trends' by focusing on citation-based influence rather than litigation or trends.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides clear context for usage by specifying that it returns 'at most Top 10 patents' and requires 'either keywords or IPC' for a meaningful search, with IPC prioritized if both are provided. However, it does not explicitly state when to use this tool versus alternatives like 'get_top_assignees' or 'get_word_cloud'.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/KunihiroS/patsnap-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server