OODA Computer Control

99 tools. 60 can modify or destroy data without limits.

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

Last updated:

60 can modify or destroy data
39 read-only
99 tools total

Community server · catalogue entry verified 12/06/2026

How to control OODA Computer Control ↓

What OODA Computer Control exposes to your agents

Read (39) Write / Execute (50) Destructive / Financial (10)
Critical Risk

The most dangerous OODA Computer Control tools

60 of OODA Computer Control's 99 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

How to control OODA Computer Control

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

Deny destructive operations
{
  "batch_delete_files": {
    "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_diff": {
    "limits": [
      {
        "counter": "apply_diff_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "start_search": {
    "limits": [
      {
        "counter": "start_search_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

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

  1. Create a free account and register OODA Computer Control — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON OODA COMPUTER CONTROL →

Free to start. No card required.

All 99 OODA Computer Control tools

EXECUTE 32 tools
Execute batch_exec_cli Execute multiple shell commands in parallel. Execute batch_keyboard_actions Execute sequence of keyboard actions (type, press, shortcut, wait). Execute batch_mouse_actions Execute sequence of mouse actions (move, click, drag, scroll, wait). Execute evaluate_js Execute JavaScript code in the page context. Execute exec_cli Execute shell commands on the host system (YOLO mode) Execute execute_code Execute code in memory without saving to file. Supports python, node, r, powershell, bash. Execute keyboard_shortcut Execute keyboard shortcut (e.g., Execute launch_application Launch an application by path or name. Execute launch_browser Launch a browser instance (Puppeteer or Playwright). Toggles headless mode. Execute navigate_page Navigate to a URL and wait for load. Execute start_process Start a new interactive process session. Returns a sessionId for subsequent interactions. Use for long-running Execute stop_search Stop a search session and cleanup resources. Execute wait Wait/sleep for specified milliseconds. Use in action sequences. Execute wait_for_window Wait for a window to appear. Matches process name or window title. Execute focus_window Bring a window to the foreground by title or PID. Execute maximize_window Maximize the active or specified window. Execute minimize_window Minimize a window or all windows. Execute notify Show a system notification. Execute click_element Click an element identified by CSS/XPath selector. Execute close_browser Close the browser instance and cleanup. Execute interact_with_process Send input to a running process session. Input is written to the process stdin. Execute keyboard_press Press a key with optional modifiers (ctrl, alt, shift). Execute keyboard_type Type text as keyboard input. Execute mouse_click Click mouse button at position. Supports double-click. Execute mouse_drag Drag from one position to another. Execute mouse_move Move mouse cursor to coordinates. Execute mouse_scroll Scroll mouse wheel. Execute move_window Move the active or specified window. Execute resize_window Resize the active or specified window. Execute restore_window Restore a minimized/maximized window. Execute set_environment Set an environment variable. Execute type_text Type text into an input field.
WRITE 18 tools
Write apply_diff Apply multiple search/replace operations to a file in a single atomic operation. Validates all blocks before a Write batch_copy_files Copy multiple files in parallel. Write batch_edit_blocks Apply multiple search/replace operations to a single file sequentially. Each edit operates on the result of th Write batch_move_files Move multiple files in parallel. Write batch_str_replace Replace strings across multiple files in parallel. Supports replaceAll option to replace multiple occurrences Write batch_write_files Write multiple files in parallel. Write clipboard_write Write text or HTML to clipboard. Write copy_file Copy a file or directory. For multiple operations, use batch_copy_files. Write crud_batch_create Create multiple records in parallel. Write crud_batch_update Update multiple records in parallel. Write crud_create Create a new record in a collection Write crud_update Update an existing record Write edit_block Search and replace text in a file with fuzzy matching fallback. Shows diff preview when exact match fails. Use Write move_file Move/rename a file or directory. For multiple operations, use batch_move_files. Write set_config_value Set a configuration value using dot notation (e.g., Write str_replace Replace a unique string in a file with another string. The string to replace must appear exactly once in the f Write write_file Write content to a file Write write_from_line Replace content starting from a specific line number. Use startLine to keep lines 1-(startLine-1) and replace
READ 39 tools
Read start_search Start a paginated file search. Returns searchId for retrieving results. Use for large directories. Read wait_for_screen_change Wait until screen content changes in a region. Useful for detecting UI updates. Read batch_file_info Get info for multiple files in parallel. Read batch_list_directories List multiple directories in parallel. Read batch_read_files Read multiple files in parallel. Read batch_search_in_files Search for patterns across multiple files in parallel. Supports regex, literal, and fuzzy/approximate matching Read clipboard_has_format Check if clipboard contains a specific format. Read clipboard_read Read clipboard contents (text, HTML, or image as base64). Read crud_batch_read Read multiple records in parallel. Read crud_query Query records in a collection Read crud_read Read a record by ID Read file_info Get file/directory metadata (size, dates, type). For multiple paths, use batch_file_info. Read find_on_screen Find text or image on screen (requires OCR/template matching dependencies). Read get_active_window Get information about the currently focused window. Read get_audit_log_stats Get audit log statistics including total entries and database size. Read get_config Get current MCP server configuration. Optionally specify a section (storage, cliPolicy, crud). Read get_console_logs Retrieve captured console logs from the browser. Read get_diff_preview Generate a diff preview showing what changes would be made without applying them. Supports unified, inline (ch Read get_environment Get environment variable(s). Read get_mouse_position Get current mouse cursor position. Read get_network_info Get network interface information. Read get_page_content Get page content in HTML, text, or markdown format. Read get_recent_tool_calls Get recent tool call history from the audit log. Useful for debugging. Read get_screen_info Get display/monitor information (resolution, count, positions). Read get_search_results Get paginated results from a search session. Automatically advances cursor for next call. Read get_system_info Get system information (OS, CPU, memory, uptime). Read get_usage_stats Get tool usage statistics including call counts, error rates, and hourly distribution. Read list_active_searches List all active search sessions with their status. Read list_directory List contents of a directory Read list_processes List running processes with CPU/memory usage. Read list_sessions List all active process sessions with their status and basic info. Read list_windows List all open windows with titles and process info. Read read_file Read file contents. ⚠️ CONTEXT WARNING: Truncates at 500 lines. For large files or targeted access, PREFER the Read read_file_lines Read specific lines from a file (token-efficient). Returns line range with optional line numbers. Use this ins Read read_process_output Read output from a process session. Use negative lines value to read last N lines. Use clear=true to clear the Read screenshot Capture screenshot of screen or region. Returns base64 image or saves to file. Read screenshot_page Capture a full-page screenshot (returns base64). Read search_files Search for files by pattern in a directory tree. Read search_in_file Search for text or regex patterns within a file. Returns matching lines with optional context. More efficient

Related servers

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

Questions about OODA Computer Control

Can an AI agent delete data through the OODA Computer Control MCP server? +

Yes. The OODA Computer Control server exposes 10 destructive tools including batch_delete_files, clear_old_logs, clipboard_clear. 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 OODA Computer Control? +

The OODA Computer Control server has 18 write tools including apply_diff, batch_copy_files, batch_edit_blocks. 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 OODA Computer Control.

How many tools does the OODA Computer Control MCP server expose? +

99 tools across 4 categories: Destructive, Execute, Read, Write. 39 are read-only. 60 can modify, create, or delete data.

How do I enforce a policy on OODA Computer Control? +

Register the OODA Computer Control 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 OODA Computer Control tool call.

Deterministic rules across all 99 OODA Computer Control tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Free to start. No card required.

99 OODA Computer Control tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.