Skip to main content
Glama

search

Search Notion workspaces for pages and databases using query filters to find specific content within your workspace.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
queryNoThe search query string
filter_object_typeNoFilter by object type
page_sizeNoNumber of results to return (max 100)

Implementation Reference

  • The main execution logic for the MCP 'search' tool: prepares params from user input, invokes NotionService.search, formats results as text content or error response.
    async ({ query, filter_object_type, page_size }) => { const params: any = { page_size }; if (query) params.query = query; if (filter_object_type) { params.filter = { property: "object", value: filter_object_type, }; } try { const results = await this.notionService.search(params); return { content: [ { type: "text", text: JSON.stringify(results, null, 2), }, ], }; } catch (error) { console.error("Error in search tool:", error); return { content: [ { type: "text", text: `Error: Failed to search Notion - ${ (error as Error).message }`, }, ], isError: true, }; } }
  • Zod input validation schema for the 'search' tool parameters.
    { query: z.string().optional().describe("The search query string"), filter_object_type: z .enum(["page", "database"]) .optional() .describe("Filter by object type"), page_size: z .number() .min(1) .max(100) .optional() .describe("Number of results to return (max 100)"), },
  • Full registration of the 'search' tool using this.server.tool(), including schema and handler function.
    private registerSearchTool(): void { this.server.tool( "search", { query: z.string().optional().describe("The search query string"), filter_object_type: z .enum(["page", "database"]) .optional() .describe("Filter by object type"), page_size: z .number() .min(1) .max(100) .optional() .describe("Number of results to return (max 100)"), }, async ({ query, filter_object_type, page_size }) => { const params: any = { page_size }; if (query) params.query = query; if (filter_object_type) { params.filter = { property: "object", value: filter_object_type, }; } try { const results = await this.notionService.search(params); return { content: [ { type: "text", text: JSON.stringify(results, null, 2), }, ], }; } catch (error) { console.error("Error in search tool:", error); return { content: [ { type: "text", text: `Error: Failed to search Notion - ${ (error as Error).message }`, }, ], isError: true, }; } } ); }
  • Helper method in NotionService that calls the underlying Notion Client search API and handles errors.
    async search(params: SearchQuery) { try { return await this.client.search(params); } catch (error) { this.handleError(error); } }
  • Zod schema and TypeScript type definition for SearchQuery parameters used by the NotionService.search helper.
    export const SearchQuerySchema = z.object({ query: z.string().optional(), filter: z .object({ value: z.enum(["page", "database"]), property: z.literal("object"), }) .optional(), sort: z .object({ direction: z.enum(["ascending", "descending"]), timestamp: z.enum(["last_edited_time"]), }) .optional(), page_size: z.number().min(1).max(100).optional(), }); export type SearchQuery = z.infer<typeof SearchQuerySchema>;

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/ramidecodes/mcp-server-notion'

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