// SCAN REPORT

Zulip

27 tools. 13 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 12 Jun 2026

13 can modify or destroy data

14 read-only

27 tools total

Community server · catalogue entry verified 12/06/2026

How to control Zulip ↓

TOOL MAP

What Zulip exposes to your agents

Read (14) Write / Execute (9) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Zulip tools

Destructive · Low remove_reaction Removing a reaction deletes existing data (the emoji reaction) and cannot be trivially undone without the original actor re-adding it. Write · Low add_reaction This tool creates new data (a reaction entry) but does not delete, execute arbitrary operations, move money, or cause irreversible damage. Write · Medium edit_message This tool modifies message content in Zulip, fitting the Write category (creates or modifies data reversibly). Write · Medium move_messages Moving messages changes their organizational state (stream/topic assignment) and is reversible, placing it in Write rather than Destructive.

13 of Zulip's 27 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Zulip

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

Deny destructive operations

{
  "remove_reaction": {
    "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

{
  "reply": {
    "limits": [
      {
        "counter": "reply_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "fetch_file": {
    "limits": [
      {
        "counter": "fetch_file_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 Zulip — 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 ZULIP →

Free to start. No card required.

FULL CATALOGUE

All 27 Zulip tools

DESTRUCTIVE 1 tools

Destructive remove_reaction Remove an emoji reaction from a message.

WRITE 9 tools

Write reply Reply in the current session context. Write add_reaction Add an emoji reaction to a message. Write edit_message edit_message Write move_messages move_messages Write resolve_topic resolve_topic Write send_direct_message Send a direct message (DM) to one or more users. Write send_message Send a message to a specific stream and topic (fire-and-forget). Write set_context set_context Write upload_file upload_file

READ 14 tools

Read fetch_file Fetch any file from Zulip and save it locally. Read fetch_image Fetch an image from Zulip and save it to a temp file for viewing. Read get_message_by_id Get a specific message by its ID. Read get_message_link get_message_link Read get_messages get_messages Read get_stream_members Get the members of a stream/channel. Read get_stream_topics Get recent topics in a stream. Read get_subscribed_streams Get streams the bot is subscribed to. Read get_user_info Get information about a Zulip user, including their full profile. Read list_emoji Search custom emoji available on this Zulip server. Read list_streams list_streams Read listen Wait for new messages in the current conversation (blocking). Read verify_message verify_message Read resolve_name resolve_name

OTHER 3 tools

Other stop_typing stop_typing Other end_session end_session Other typing typing

EXPLORE

Related servers

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

FAQ

Questions about Zulip

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

Yes. The Zulip server exposes 1 destructive tools including remove_reaction. 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 Zulip? +

The Zulip server has 9 write tools including reply, add_reaction, edit_message. 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 Zulip.

How many tools does the Zulip MCP server expose? +

27 tools across 4 categories: Destructive, Execute, Read, Write. 14 are read-only. 13 can modify, create, or delete data.

How do I enforce a policy on Zulip? +

Register the Zulip 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 Zulip tool call.

Deterministic rules across all 27 Zulip tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON ZULIP →

Free to start. No card required.

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