// SCAN REPORT

CustomGPT MCP Server

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

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

Last updated: 29 Jun 2026

24 can modify or destroy data

26 read-only

50 tools total

Community server · catalogue entry verified 29/06/2026

How to control CustomGPT MCP Server ↓

TOOL MAP

What CustomGPT MCP Server exposes to your agents

Read (26) Write / Execute (19) Destructive / Financial (5)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous CustomGPT MCP Server tools

Destructive · High delete_agent Deletion of an agent is a destructive operation that permanently removes data and configurations. Destructive · High delete_conversation This tool permanently removes data (a conversation and all associated messages) that cannot be recovered or undone. Destructive · High delete_license Deleting a license irreversibly removes access credentials and licensing agreements. Destructive · High delete_page The tool performs an irreversible deletion operation on a page resource within an agent.

24 of CustomGPT MCP Server's 50 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control CustomGPT MCP Server

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

Deny destructive operations

{
  "delete_agent": {
    "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

{
  "create_agent": {
    "limits": [
      {
        "counter": "create_agent_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "get_agent": {
    "limits": [
      {
        "counter": "get_agent_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 CustomGPT MCP 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 CUSTOMGPT →

Instant setup, no code required.

FULL CATALOGUE

All 50 CustomGPT MCP Server tools

DESTRUCTIVE 5 tools

Destructive delete_agent Delete a CustomGPT agent. Destructive delete_conversation Delete a conversation and all its messages. Destructive delete_license Delete a license. Destructive delete_page Delete a specific page from an agent. Destructive delete_source Delete a source from an agent.

EXECUTE 3 tools

Execute send_message Send a message to a CustomGPT agent. Execute stream_to_claude Stream conversation to Claude via CustomGPT with timeout prevention Execute synchronize_source Synchronize/refresh a source.

WRITE 16 tools

Write create_agent Create a new CustomGPT agent. Write create_agent_license Create a new license for an agent. Write create_conversation Create a new conversation for an agent. Write create_plugin Create a plugin for an agent. Write create_source Create a new source for an agent. Write replicate_agent Replicate/clone an agent by copying all its info, settings, sources and files. Write send_conversation_message Send a message to a specific conversation session. Write update_agent Update an agent's basic information. Write update_agent_settings Update comprehensive agent settings with all available configuration options. Write update_conversation Update a conversation's name. Write update_license Update a license name. Write update_message_feedback Update feedback reaction for a specific message (thumbs up/down). Write update_page_metadata Update metadata for a specific page. Write update_plugin Update a plugin for an agent. Write update_source_settings Update settings for a specific source (sitemap or upload). Write update_user_profile Update user profile.

READ 26 tools

Read get_agent Get details for a specific agent. Read get_agent_settings Get configuration settings for an agent. Read get_agent_stats Get statistics for a CustomGPT agent. Read get_analysis_report Get graph-ready analysis data with various metrics (queries, conversations, queries per conversation). Read get_citation Get citation details. Read get_conversation_messages Get all messages in a specific conversation. Read get_conversations_report Get conversations analytics for an agent. Read get_intelligence_report Get customer intelligence analytics data including user interactions, emotions, intents, and behavioral analyt Read get_license_details Get details for a specific license. Read get_message_details Get detailed information for a specific message by ID. Read get_page_metadata Get metadata for a specific page. Read get_queries_report Get queries analytics for an agent. Read get_server_info Get server information and available tools. Read get_traffic_report Get traffic analytics for an agent. Read get_usage_limits Get account usage limits. Read get_user_profile Get user profile information. Read list_agent_licenses List all licenses for an agent. Read list_agents List all your CustomGPT agents. Read list_conversations List all conversations for a specific agent. Read list_pages List all pages for an agent. Read list_plugins List plugins for an agent. Read list_sources List all sources (sitemaps and files) for an agent. Read preview_page Preview a file from citation using preview ID. Read reindex_page Reindex a specific page to refresh its content. Read search_team_member Search for a team member by ID or email (team owners/admins only). Read validate_api_key Validate the configured CustomGPT API key.

EXPLORE

Related servers

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

FAQ

Questions about CustomGPT MCP Server

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

Yes. The CustomGPT MCP Server server exposes 5 destructive tools including delete_agent, delete_conversation, delete_license. 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 CustomGPT MCP Server? +

The CustomGPT MCP Server server has 16 write tools including create_agent, create_agent_license, create_conversation. 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 CustomGPT MCP Server.

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

50 tools across 3 categories: Destructive, Read, Write. 26 are read-only. 24 can modify, create, or delete data.

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

Register the CustomGPT MCP 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 CustomGPT MCP Server tool call.

Deterministic rules across all 50 CustomGPT MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON CUSTOMGPT →

Instant setup, no code required.

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