Skip to main content
Glama

read_file

Read file contents from a specified path with configurable text encoding options for processing and automation workflows.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pathYes
encodingNoutf8

Implementation Reference

  • The execution handler for the 'read_file' MCP tool. Validates the file path existence, reads the file contents asynchronously using Node.js fs.promises.readFile, retrieves file statistics, and returns a structured response containing the text content and metadata including path, size, and encoding.
    async ({ path: filePath, encoding }) => {
      return wrapToolExecution(async () => {
        const validatedPath = await validatePathExists(filePath);
        const content = await fs.readFile(validatedPath, { encoding: encoding as FileEncoding });
        const stats = await fs.stat(validatedPath);
    
        return {
          content: [{ type: "text" as const, text: content }],
          metadata: {
            path: validatedPath,
            size: stats.size,
            encoding
          }
        };
      }, {
        errorCode: ERROR_CODES.FILE_OPERATION,
        context: "Failed to read file"
      });
    }
  • Zod input schema definition for the 'read_file' tool: 'path' is a required non-empty string, 'encoding' is an optional enum (utf8, utf-16le, latin1) defaulting to the configured default.
    {
      path: z.string().min(1, "Path is required"),
      encoding: z.enum(["utf8", "utf-16le", "latin1"]).optional().default(DEFAULTS.FILE_ENCODING)
    },
  • The registerReadFile function registers the 'read_file' tool on the MCP server instance using server.tool(), providing the name, Zod schema, and handler function.
    function registerReadFile(server: McpServer): void {
      server.tool("read_file",
        {
          path: z.string().min(1, "Path is required"),
          encoding: z.enum(["utf8", "utf-16le", "latin1"]).optional().default(DEFAULTS.FILE_ENCODING)
        },
        async ({ path: filePath, encoding }) => {
          return wrapToolExecution(async () => {
            const validatedPath = await validatePathExists(filePath);
            const content = await fs.readFile(validatedPath, { encoding: encoding as FileEncoding });
            const stats = await fs.stat(validatedPath);
    
            return {
              content: [{ type: "text" as const, text: content }],
              metadata: {
                path: validatedPath,
                size: stats.size,
                encoding
              }
            };
          }, {
            errorCode: ERROR_CODES.FILE_OPERATION,
            context: "Failed to read file"
          });
        }
      );
    }
  • src/index.ts:65-65 (registration)
    Higher-level registration call to registerFileTools(server), which internally calls registerReadFile to add the 'read_file' tool among file-related tools.
    registerFileTools(server);

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/ishuru/open-mcp'

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