About trash-cleaner-mcp-server
A desktop MCP service for scanning and cleaning junk files, primarily designed and tested for macOS. It also provides filesystem analysis, VS Code extension cleanup, audit logs, and scheduled tool execution.
Key features
- 01Scan files and analyze directory sizes
- 02Clean caches, temporary files, and app remnants
- 03Find and clean outdated VS Code extensions
- 04Inspect files and perform confirmed deletion
- 05Log operations and schedule tools
Use cases
- 01Find large files consuming disk space
- 02Remove old application caches and temporary files
- 03Identify outdated VS Code extensions
- 04Schedule recurring cleanup tasks
MCP support
Exposes cleanup, filesystem, system information, audit, and scheduler tools to MCP clients. Most system-cleaning tools are designed and tested specifically for macOS.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
25Saved tool list · updated 3w ago
This tool takes no parameters.
Details
From the project READMEtrash-cleaner-mcp-server
A desktop MCP service providing tools to scan, analyze, and clean junk files, primarily optimized for macOS.
Note: Most cleaning tools (cleanAppCaches, cleanTempFiles, cleanAppRemnants, smartCleanSystem) are designed and tested specifically for macOS. Running them on other systems might not work or lead to unexpected behavior.
Features
- Cross-Platform: (Planned) Windows, macOS, Linux support.
- Intelligent Scanning: Identifies various types of junk files (caches, logs, temp files, etc.).
- Selective Cleaning: Tools for cleaning specific areas like application caches, temp files, and application remnants.
- VS Code Extension Management: Tools to find and clean old/unused VS Code extensions.
- Filesystem Utilities: Tools for directory analysis, file finding, size calculation, deletion, etc.
- System Information: Basic OS detection.
- Audit Logging: Tracks operations via
[Audit]tools. - Scheduled Tasks: Automate cleaning and other tasks using a flexible scheduler.
- Define tasks using cron expressions or simple intervals.
- Target any registered tool for execution.
- Create/update tasks using exact tool names (
toolName) or natural language queries (toolQuery) via fuzzy matching. - Full task management (list, details, update, enable/disable, delete).
- Manual triggering and execution history tracking.
Available Tools (Features)
This server registers several tools callable via the MCP protocol:
macOS Cleaning Tools (Optimized for macOS):
macOSWarning: Displays a warning if run on a non-macOS system.cleanAppCaches: Cleans various application caches (user cache, saved app state).- Options:
dryRun(bool, default: true),olderThan(int, optional - days).
- Options:
cleanTempFiles: Cleans system temporary files.- Options:
dryRun(bool, default: true),olderThan(int, optional, default: 7 days).
- Options:
cleanAppRemnants: Scans for and cleans configuration/cache files left by uninstalled applications.- Options:
appName(string, optional - partial name match),dryRun(bool, default: true).
- Options:
smartCleanSystem: Performs system cleaning based on predefined levels (targets common junk locations).- Options:
cleanLevel(enum: "safe", "normal", "deep", default: "safe"),dryRun(bool, default: true).
- Options:
Cross-Platform Tools (Scanning, File System, OS):
cleanVSCodeExtensions: Cleans outdated versions of VS Code (and compatible editors like VSCodium, Cursor) extensions.- Options:
editorPath(string, optional - path to specific editor's extensions, default: scans all known locations),dryRun(bool, default: true).
- Options:
reportVSCodeExtensions: Generates a report listing outdated VS Code extensions that can be cleaned.- Options:
editorPath(string, optional).
- Options:
findLargeFiles: Finds files exceeding a specified size within a directory.- Options:
path(string),minSize(int, default: 100MB),maxDepth(int, default: 3).
- Options:
scanFullSystem: Scans the entire system's file structure (respecting exclusions) to report on space usage.- Options:
excludePaths(string[], optional),maxDepth(int, default: 10).
- Options:
scanDirectory: Analyzes the contents and size distribution within a specific directory.- Options:
path(string),excludePaths(string[], optional),maxDepth(int, default: 10),includeSubdirs(bool, default: true).
- Options:
getFolderSize: Calculates and returns the total size of a folder.- Options:
path(string).
- Options:
listDirectory: Lists the files and subdirectories within a specified path.- Options:
path(string),showHidden(bool, default: false).
- Options:
findFiles: Searches for files matching a specific pattern (wildcards supported) within a directory.- Options:
path(string),pattern(string),maxDepth(int, default: 5).
- Options:
analyzeDirectory: Provides statistics on file types and size distribution within a directory.- Options:
path(string),includeSubdirs(bool, default: true).
- Options:
getFileHash: Computes the MD5, SHA1, or SHA256 hash of a file.- Options:
path(string),algorithm(enum: "md5", "sha1", "sha256", default: "md5").
- Options:
getFileType: Retrieves basic file information, including MIME type.- Options:
path(string).
- Options:
deletePath: Deletes a specified file or folder (requires explicit confirmation, with extra checks for dangerous paths).- Options:
path(string),confirm(bool, must be true),dangerConfirm(bool, optional - required for potentially risky deletions).
- Options:
checkPathExists: Checks if a given path exists and identifies if it's a file or directory.- Options:
path(string).
- Options:
getSystemType: Returns the operating system identifier (e.g., 'darwin', 'win32', 'linux').
Audit Tools:
viewAuditLog: Views the most recent lines from the application's main log file (combined.log).- Options:
lines(int, default: 100, max: 1000).
- Options:
clearAuditLog: Deletes the main application log file (combined.log). Requires confirmation.- Options:
confirm(bool, must be true).
- Options:
Scheduled Tasks (TrashCleaner_Scheduler Tools)
Provides functionality to schedule the execution of other registered tools.
TrashCleaner_Scheduler_CreateTask: Creates a new scheduled task.- Requires
name,cronExpression. - Requires either
toolName(exact tool name like"TrashCleaner_CleanAppCaches") ortoolQuery(fuzzy description like"clean app caches"). - Optional
toolParams,enabled.
- Requires
TrashCleaner_Scheduler_ListTasks: Lists all configured scheduled tasks and their status.TrashCleaner_Scheduler_GetTaskDetails: Gets detailed information about a specific task by ID.TrashCleaner_Scheduler_UpdateTask: Updates an existing task. Can modify name, schedule, target tool (viatoolNameortoolQuery), parameters, and enabled status.TrashCleaner_Scheduler_EnableTask: Enables a disabled task.TrashCleaner_Scheduler_DisableTask: Disables an enabled task.TrashCleaner_Scheduler_DeleteTask: Permanently deletes a scheduled task.TrashCleaner_Scheduler_RunTaskNow: Manually triggers the execution of a task immediately.TrashCleaner_Scheduler_GetTaskHistory: Retrieves the recent execution history (success/failure) for a task.
Tech Stack
- Bun.js
- TypeScript
Usage
After publishing to npm, you can run the MCP server using either Bun or npm:
Using Bun:
bunx @childhoodandy/trash-cleaner-mcp-server
# or with auto-install
bunx -y @childhoodandy/trash-cleaner-mcp-server
Using npm:
npx @childhoodandy/trash-cleaner-mcp-server
# or with auto-install
npx -y @childhoodandy/trash-cleaner-mcp-server
Installation
# Clone the repository
git clone <repository-url>
cd trash-cleaner-mcp-server
# Install dependencies (using Bun)
bun install
Running the Server
bun start
Configuration
- Logging configuration is in
src/utils/logger.ts. - Scheduled task definitions are stored in
~/.trash-cleaner/schedules.json.
Development
(Development guidelines remain here...)
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.