Zulip MCP Server
Related Servers
Alternatives to Zulip MCP Server
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityBmaintenanceBridges MCP-compatible AI assistants to Zulip chat, enabling message sending/receiving, search, user resolution, and dual-identity support.33MIT
- FlicenseNot gradedqualityDmaintenanceMCP server that exposes Zulip REST API tools via SSE, enabling message retrieval, stream/topic listing, draft management, and notifications.-
- AlicenseAqualityDmaintenanceRun AI agents in Zulip as @mentionable bots — or wire into any MCP client.2729Apache 2.0
- AlicenseAqualityBmaintenanceEnables MCP clients to log in to a self-hosted chat instance as a DSM user, list channels and direct messages, read recent posts, find users, send text messages, and upload files.6MIT
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients to list Slack channels, post and reply to messages, add reactions, and retrieve channel history, thread replies, users, and user profiles through the Slack Web API.74 npmMIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents that speak MCP to obtain identities, exchange direct messages and threaded replies, check inboxes, long-poll for new messages, and manage agent access across a shared hub.MIT
TDQS
Scored across 9 tools
Most tools target a clearly distinct resource+action: channel listing, channel posting, DM sending, reactions, topics, subscriptions, and users are all separable. The only mild overlap is between zulip_get_channel_history and zulip_get_my_mentions, which both return messages but with different scopes (single channel/topic vs. cross-channel mentions/DMs), and descriptions make that distinction clear.
All nine tools follow a uniform zulip_<verb>_<noun> snake_case pattern (zulip_list_channels, zulip_post_message, zulip_get_topics, etc.). Verb choices vary (post/send/add/list/get) but each verb fits its object naturally rather than reflecting inconsistency.
Nine tools is well within the ideal 3-15 range for a chat-platform integration and each one maps to a distinct Zulip primitive. No tool feels redundant or like padding.
Core read/write flows are covered (list channels, post to channel, send DM, topics, history, mentions, users), but the message lifecycle is incomplete: there is no edit/update message, delete message, or remove-reaction operation, and no way to fetch direct-message history with a specific user. Agents wanting to correct or clean up messages will hit dead ends.