// SCAN REPORT

Grok MCP

22 tools. 12 can modify or destroy data without limits.

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

Last updated: 11 Jun 2026

12 can modify or destroy data

10 read-only

22 tools total

Community server · catalogue entry verified 11/06/2026

How to control Grok MCP ↓

TOOL MAP

What Grok MCP exposes to your agents

Read (10) Write / Execute (9) Destructive / Financial (3)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Grok MCP tools

Destructive · High clear_chat_history This tool deletes chat history data without the ability to recover it. Destructive · High delete_file This tool irreversibly deletes data (an uploaded file) with no undo capability. Destructive · High delete_stateful_response This tool permanently removes data (chat completion records) from XAI's servers without possibility of recovery. Execute · High generate_image Based on the tool name and server context ('image understanding and generation'), this tool likely calls an external AI API to generate images.

12 of Grok MCP's 22 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Grok MCP

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

Deny destructive operations

{
  "clear_chat_history": {
    "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

{
  "stateful_chat": {
    "limits": [
      {
        "counter": "stateful_chat_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "chat_with_vision": {
    "limits": [
      {
        "counter": "chat_with_vision_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 Grok MCP — 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 GROK →

Free to start. No card required.

FULL CATALOGUE

All 22 Grok MCP tools

DESTRUCTIVE 3 tools

Destructive clear_chat_history clear_chat_history Destructive delete_file Permanently delete an uploaded file from xAI. Destructive delete_stateful_response Delete a stored chat completion from xAI's servers.

EXECUTE 7 tools

Execute chat chat Execute chat_with_files chat_with_files Execute code_executor code_executor Execute extend_video extend_video Execute grok_agent grok_agent Execute generate_image generate_image Execute generate_video generate_video

WRITE 2 tools

Write stateful_chat stateful_chat Write upload_file upload_file

READ 10 tools

Read chat_with_vision chat_with_vision Read get_chat_history Return the full message history for a local chat session. Read get_file Fetch metadata for a single uploaded file. Read get_file_content get_file_content Read list_chat_sessions List all local chat sessions stored under chats/. Read list_files list_files Read list_models List all Grok language and image models with live pricing from the xAI API. Read retrieve_stateful_response Fetch a stored chat completion from xAI by its response ID. Read web_search web_search Read x_search x_search

EXPLORE

Related servers

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

FAQ

Questions about Grok MCP

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

Yes. The Grok MCP server exposes 3 destructive tools including clear_chat_history, delete_file, delete_stateful_response. 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 Grok MCP? +

The Grok MCP server has 2 write tools including stateful_chat, upload_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 Grok MCP.

How many tools does the Grok MCP server expose? +

22 tools across 3 categories: Destructive, Read, Write. 10 are read-only. 12 can modify, create, or delete data.

How do I enforce a policy on Grok MCP? +

Register the Grok 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 Grok MCP tool call.

Deterministic rules across all 22 Grok MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON GROK →

Free to start. No card required.

22 Grok MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.