// SCAN REPORT

MCP Workspace Server

36 tools. 24 can modify or destroy data without limits.

2 destructive tools with no built-in limits. Policy required.

Last updated: 4 Jul 2026

24 can modify or destroy data

12 read-only

36 tools total

Community server · catalogue entry verified 04/07/2026

How to control MCP Workspace Server ↓

TOOL MAP

What MCP Workspace Server exposes to your agents

Read (12) Write / Execute (22) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous MCP Workspace Server tools

Destructive · High cloud_storage Because the tool bundles delete alongside upload/download/list, and deletion of cloud storage files can be irreversible, the most severe applicable category is Destructive. Destructive · High delete_file This tool irreversibly removes data (files or directories) from the workspace. Execute · High run_command This tool runs commands, making it Execute-category by definition. Execute · High docker_manage Docker management encompasses executing containers (running arbitrary code/processes), controlling networks, and managing volumes.

24 of MCP Workspace Server's 36 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control MCP Workspace Server

PolicyLayer is an MCP gateway — it sits between your AI agents and MCP Workspace Server, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "cloud_storage": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "apply_patch": {
    "limits": [
      {
        "counter": "apply_patch_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "find_files": {
    "limits": [
      {
        "counter": "find_files_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register MCP Workspace Server — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON MCP WORKSPACE →

Instant setup, no code required.

FULL CATALOGUE

All 36 MCP Workspace Server tools

DESTRUCTIVE 2 tools

Destructive cloud_storage Interact with cloud storage services - upload, download, list, delete files Destructive delete_file Delete a file or empty directory from the workspace. Non-empty directories cannot be deleted.

EXECUTE 11 tools

Execute code_format Format, lint, and analyze code using various formatting tools Execute database_query Execute SQL queries on SQLite databases within the workspace Execute docker_manage Manage Docker containers, images, networks, and volumes Execute encrypt_decrypt Encrypt/decrypt files or text, generate hashes, create encryption keys Execute git_branch List, create, switch, and manage git branches Execute http_request Make HTTP requests to external APIs and services Execute kill_process Terminate running processes by PID or name Execute package_manager Manage packages with npm, yarn, pip, composer, and other package managers Execute run_command Execute an allowed command in the workspace. Only commands in the allowlist can be executed. Execute schedule_task Schedule commands to run at specific times using system task scheduler Execute text_process Advanced text processing - analyze, transform, extract patterns, compare, generate text

WRITE 11 tools

Write apply_patch Apply a patch to an existing file in the workspace. Uses a simple patch format: <<<OLD\\n...\\n===\\n...\\n>>> Write compress_files Create compressed archives (ZIP, TAR, etc.) from files and directories Write copy_file Copy files or directories from one location to another Write create_folder Create a directory in the workspace. Automatically creates parent directories if they do not exist. Idempotent Write csv_process Process CSV files - read, write, transform, analyze, filter, merge, split Write extract_archive Extract compressed archives (ZIP, TAR, GZ, BZ2, etc.) Write image_process Process images - resize, crop, rotate, convert formats, get info Write move_file Move or rename files and directories Write pdf_manipulate Extract text from PDFs, get info, merge, or split PDF files Write send_notification Send system notifications, webhooks, or email alerts Write write_file Create or overwrite a file in the workspace. Supports atomic writes and automatic parent directory creation.

READ 12 tools

Read find_files Find files by name pattern, extension, or other criteria Read get_file_info Get detailed information about files and directories Read git_diff Show differences between commits, branches, or working directory Read git_log Show git commit history with detailed information Read git_status Get git repository status and changes Read json_process Process JSON data - parse, validate, transform, merge, extract, format Read list_files List files and directories in the workspace. Returns file metadata including name, path, type, size, and last Read list_processes List running processes with CPU, memory, and other details Read read_file Read file contents from the workspace. Returns file content as UTF-8 text along with metadata including path, Read search_files Search for text content within files in the workspace Read system_info Get system information and resource usage Read web_scrape Scrape web pages - fetch HTML, extract elements, get links, images, text, metadata

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about MCP Workspace Server

Can an AI agent delete data through the MCP Workspace Server MCP server? +

Yes. The MCP Workspace Server server exposes 2 destructive tools including cloud_storage, delete_file. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through MCP Workspace Server? +

The MCP Workspace Server server has 11 write tools including apply_patch, compress_files, copy_file. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach MCP Workspace Server.

How many tools does the MCP Workspace Server MCP server expose? +

36 tools across 4 categories: Destructive, Execute, Read, Write. 12 are read-only. 24 can modify, create, or delete data.

How do I enforce a policy on MCP Workspace Server? +

Register the MCP Workspace Server MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every MCP Workspace Server tool call.

Deterministic rules across all 36 MCP Workspace Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MCP WORKSPACE →

Instant setup, no code required.

36 MCP Workspace Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.