Skip to main content
Glama

cleanup_data

Removes all NotebookLM MCP server data files across 8 categories including installations, caches, logs, and browser sessions. Shows preview before deletion and preserves library if specified.

Instructions

ULTRATHINK Deep Cleanup - Scans entire system for ALL NotebookLM MCP data files across 8 categories. Always runs in deep mode, shows categorized preview before deletion.

āš ļø CRITICAL: Close ALL Chrome/Chromium instances BEFORE running this tool! Open browsers can prevent cleanup and cause issues.

Categories scanned:

  1. Legacy Installation (notebooklm-mcp-nodejs) - Old paths with -nodejs suffix

  2. Current Installation (notebooklm-mcp) - Active data, browser profiles, library

  3. NPM/NPX Cache - Cached installations from npx

  4. Claude CLI MCP Logs - MCP server logs from Claude CLI

  5. Temporary Backups - Backup directories in system temp

  6. Claude Projects Cache - Project-specific cache (optional)

  7. Editor Logs (Cursor/VSCode) - MCP logs from code editors (optional)

  8. Trash Files - Deleted notebooklm files in system trash (optional)

Works cross-platform (Linux, Windows, macOS). Safe by design: shows detailed preview before deletion, requires explicit confirmation.

LIBRARY PRESERVATION: Set preserve_library=true to keep your notebook library.json file while cleaning everything else.

RECOMMENDED WORKFLOW for fresh start:

  1. Ask user to close ALL Chrome/Chromium instances

  2. Run cleanup_data(confirm=false, preserve_library=true) to preview

  3. Run cleanup_data(confirm=true, preserve_library=true) to execute

  4. Run setup_auth or re_auth for fresh browser session

Use cases: Clean reinstall, troubleshooting auth issues, removing all traces before uninstall, cleaning old browser sessions and installation data.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
confirmYesConfirmation flag. Tool shows preview first, then user confirms deletion. Set to true only after user has reviewed the preview and explicitly confirmed.
preserve_libraryNoPreserve library.json file during cleanup. Default: false. Set to true to keep your notebook library while deleting everything else (browser data, caches, logs).

Implementation Reference

  • The primary handler function that executes the cleanup_data tool. It handles both preview (confirm=false) and deletion (confirm=true) modes, instantiates CleanupManager, logs progress, and returns structured ToolResult with preview data or deletion results.
    /**
     * Handle cleanup_data tool
     *
     * ULTRATHINK Deep Cleanup - scans entire system for ALL NotebookLM MCP files
     */
    async handleCleanupData(
      args: { confirm: boolean; preserve_library?: boolean }
    ): Promise<
      ToolResult<{
        status: string;
        mode: string;
        preview?: {
          categories: Array<{ name: string; description: string; paths: string[]; totalBytes: number; optional: boolean }>;
          totalPaths: number;
          totalSizeBytes: number;
        };
        result?: {
          deletedPaths: string[];
          failedPaths: string[];
          totalSizeBytes: number;
          categorySummary: Record<string, { count: number; bytes: number }>;
        };
      }> 
    > {
      const { confirm, preserve_library = false } = args;
    
      log.info(`šŸ”§ [TOOL] cleanup_data called`);
      log.info(`  Confirm: ${confirm}`);
      log.info(`  Preserve Library: ${preserve_library}`);
    
      const cleanupManager = new CleanupManager();
    
      try {
        // Always run in deep mode
        const mode = "deep";
    
        if (!confirm) {
          // Preview mode - show what would be deleted
          log.info(`  šŸ“‹ Generating cleanup preview (mode: ${mode})...`);
    
          const preview = await cleanupManager.getCleanupPaths(mode, preserve_library);
          const platformInfo = cleanupManager.getPlatformInfo();
    
          log.info(`  Found ${preview.totalPaths.length} items (${cleanupManager.formatBytes(preview.totalSizeBytes)})`);
          log.info(`  Platform: ${platformInfo.platform}`);
    
          return {
            success: true,
            data: {
              status: "preview",
              mode,
              preview: {
                categories: preview.categories,
                totalPaths: preview.totalPaths.length,
                totalSizeBytes: preview.totalSizeBytes,
              },
            },
          };
        } else {
          // Cleanup mode - actually delete files
          log.info(`  šŸ—‘ļø  Performing cleanup (mode: ${mode})...`);
    
          const result = await cleanupManager.performCleanup(mode, preserve_library);
    
          if (result.success) {
            log.success(`āœ… [TOOL] cleanup_data completed - deleted ${result.deletedPaths.length} items`);
          } else {
            log.warning(`āš ļø  [TOOL] cleanup_data completed with ${result.failedPaths.length} errors`);
          }
    
          return {
            success: result.success,
            data: {
              status: result.success ? "completed" : "partial",
              mode,
              result: {
                deletedPaths: result.deletedPaths,
                failedPaths: result.failedPaths,
                totalSizeBytes: result.totalSizeBytes,
                categorySummary: result.categorySummary,
              },
            },
          };
        }
      } catch (error) {
        const errorMessage = error instanceof Error ? error.message : String(error);
        log.error(`āŒ [TOOL] cleanup_data failed: ${errorMessage}`);
        return {
          success: false,
          error: errorMessage,
        };
      }
    }
  • Tool definition including name, detailed description, and inputSchema specifying 'confirm' (required boolean) and 'preserve_library' (optional boolean with default false).
    {
      name: "cleanup_data",
      description:
        "ULTRATHINK Deep Cleanup - Scans entire system for ALL NotebookLM MCP data files across 8 categories. Always runs in deep mode, shows categorized preview before deletion.\n\n" +
        "āš ļø CRITICAL: Close ALL Chrome/Chromium instances BEFORE running this tool! Open browsers can prevent cleanup and cause issues.\n\n" +
        "Categories scanned:\n" +
        "1. Legacy Installation (notebooklm-mcp-nodejs) - Old paths with -nodejs suffix\n" +
        "2. Current Installation (notebooklm-mcp) - Active data, browser profiles, library\n" +
        "3. NPM/NPX Cache - Cached installations from npx\n" +
        "4. Claude CLI MCP Logs - MCP server logs from Claude CLI\n" +
        "5. Temporary Backups - Backup directories in system temp\n" +
        "6. Claude Projects Cache - Project-specific cache (optional)\n" +
        "7. Editor Logs (Cursor/VSCode) - MCP logs from code editors (optional)\n" +
        "8. Trash Files - Deleted notebooklm files in system trash (optional)\n\n" +
        "Works cross-platform (Linux, Windows, macOS). Safe by design: shows detailed preview before deletion, requires explicit confirmation.\n\n" +
        "LIBRARY PRESERVATION: Set preserve_library=true to keep your notebook library.json file while cleaning everything else.\n\n" +
        "RECOMMENDED WORKFLOW for fresh start:\n" +
        "1. Ask user to close ALL Chrome/Chromium instances\n" +
        "2. Run cleanup_data(confirm=false, preserve_library=true) to preview\n" +
        "3. Run cleanup_data(confirm=true, preserve_library=true) to execute\n" +
        "4. Run setup_auth or re_auth for fresh browser session\n\n" +
        "Use cases: Clean reinstall, troubleshooting auth issues, removing all traces before uninstall, cleaning old browser sessions and installation data.",
      inputSchema: {
        type: "object",
        properties: {
          confirm: {
            type: "boolean",
            description:
              "Confirmation flag. Tool shows preview first, then user confirms deletion. " +
              "Set to true only after user has reviewed the preview and explicitly confirmed.",
          },
          preserve_library: {
            type: "boolean",
            description:
              "Preserve library.json file during cleanup. Default: false. " +
              "Set to true to keep your notebook library while deleting everything else (browser data, caches, logs).",
            default: false,
          },
        },
        required: ["confirm"],
      },
    },
  • The buildToolDefinitions function aggregates all tool definitions, including systemTools (which contains cleanup_data), making it available to the MCP server via list_tools.
    export function buildToolDefinitions(library: NotebookLibrary): Tool[] {
      // Update the description for ask_question based on the library state
      const dynamicAskQuestionTool = {
        ...askQuestionTool,
        description: buildAskQuestionDescription(library),
      };
    
      return [
        dynamicAskQuestionTool,
        ...notebookManagementTools,
        ...sessionManagementTools,
        ...systemTools,
      ];
    }
  • src/index.ts:266-270 (registration)
    Dispatch case in the main MCP call_tool handler that routes 'cleanup_data' calls to the specific handleCleanupData method.
    case "cleanup_data":
      result = await this.toolHandlers.handleCleanupData(
        args as { confirm: boolean }
      );
      break;

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.0.0

TDQS

A4.7/5.0
Behavior5/5

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

With no annotations, the description carries the full burden of behavior disclosure. It clearly indicates destructive deletion, preview-before-deletion, explicit confirmation, deep mode, Chrome/Chromium prerequisite, cross-platform behavior, and the library preservation option. This is far beyond minimal disclosure and fully prepares an agent for the tool's safety profile.

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

Conciseness5/5

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

Although the description is long, it is well-structured with a critical warning front-loaded, a scannable categorized list, and clear sections for workflow and use cases. Every element serves the operational needs of a destructive, cross-platform tool; there is no filler or redundancy.

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

Completeness5/5

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

Given the tool's complexity, destructive nature, absence of annotations, and absence of an output schema, the description is remarkably complete. It covers prerequisites, the eight categories scanned, the preservation escape hatch, a step-by-step workflow, and use cases. No critical detail needed to invoke the tool safely is missing.

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

Parameters4/5

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

Schema coverage is 100% for both parameters, so the baseline is 3. The description adds a recommended workflow that maps confirm=false to preview and confirm=true to execution, and clarifies preserve_library as keeping library.json 'while cleaning everything else.' This usage-level guidance goes beyond the schema's parameter descriptions.

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 states a specific verb (cleanup) and resource ('all NotebookLM MCP data files across 8 categories'), with explicit scope. It is clearly differentiated from sibling tools, which are about notebooks, sessions, auth, or health, none of which perform system-wide data deletion. The 'Always runs in deep mode' detail adds further precision.

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?

It provides explicit use cases (clean reinstall, troubleshooting auth issues, removing all traces before uninstall, cleaning old browser sessions) and a recommended workflow that sequences preview and execute calls, followed by setup_auth/re_auth. It does not explicitly state when not to use the tool, but since no sibling tool is an alternative for cleanup, the guidance is sufficient.

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