weflow-mcp
Provides tools for interacting with WeChat through the local WeFlow HTTP API, enabling management of chat sessions, messages, contacts, group members, and Moments (timeline) data.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@weflow-mcp list my recent chat sessions"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
weflow-mcp
TypeScript MCP server for calling the local WeFlow HTTP API.
WeFlow must be running locally with API 服务 enabled. The default API base URL is http://127.0.0.1:5031. Except for health checks, WeFlow /api/v1/* endpoints require an access token.
Install
Use directly with npx:
npx -y weflow-mcpOr install from source:
npm install
npm run buildRelated MCP server: yan
Configuration
Set these environment variables in your MCP client:
Variable | Default | Description |
|
| WeFlow HTTP API base URL |
| empty | WeFlow API token |
| empty | Fallback token name, used if |
|
| Request timeout in milliseconds |
MCP client example
{
"mcpServers": {
"weflow": {
"command": "node",
"args": ["/home/projects/wechat-tool/weflow-mcp/dist/index.js"],
"env": {
"WEFLOW_BASE_URL": "http://127.0.0.1:5031",
"WEFLOW_ACCESS_TOKEN": "YOUR_TOKEN"
}
}
}
}After npm publish, you can use npx instead:
{
"mcpServers": {
"weflow": {
"command": "npx",
"args": ["-y", "weflow-mcp"],
"env": {
"WEFLOW_BASE_URL": "http://127.0.0.1:5031",
"WEFLOW_ACCESS_TOKEN": "YOUR_TOKEN"
}
}
}
}Tools
weflow_health: check WeFlow API health.weflow_list_sessions: list chat sessions.weflow_get_messages: get messages from a chat session.weflow_get_session_messages_chatlab: pull one session in ChatLab format.weflow_list_contacts: list contacts.weflow_get_group_members: list group members.weflow_sns_timeline: get Moments timeline.weflow_sns_usernames: list Moments publishers.weflow_sns_export_stats: get Moments export stats.weflow_sns_export: start a Moments export.weflow_sns_block_delete: get/install/uninstall Moments anti-delete status.weflow_request: call any supported WeFlow API endpoint with custom method/query/body.
Notes
WeFlow's API only listens on
127.0.0.1by default.Before using data tools, connect WeFlow to the WeChat database in the WeFlow app.
Media URLs returned by message APIs are usable after calling messages with
media=1.
Available Tools
12 toolsweflow_get_group_membersB
List members of a WeChat group from WeFlow, optionally including message counts.
Inputs: chatroomId (required), withCounts, forceRefresh, includeMessageCounts.
weflow_get_messagesC
Get messages from one WeFlow session by talker ID. Supports date ranges, keyword filtering, ChatLab format, and media export flags.
Inputs: talker (required), end, emoji, image, +9 more.
weflow_get_session_messages_chatlabB
Pull one session in ChatLab format using WeFlow's /api/v1/sessions/:id/messages endpoint.
Inputs: id (required), end, limit, since, +1 more.
weflow_healthA
Check whether the local WeFlow HTTP API is reachable.
Inputs: no named parameters.
weflow_list_contactsB
List WeFlow contacts. Use keyword to search username, nickname, remark, or display name.
Inputs: limit, keyword.
weflow_list_sessionsC
List WeFlow chat sessions. Use keyword to search username or display name.
Inputs: limit, format, keyword.
weflow_requestB
Call a custom WeFlow API endpoint. Use this for endpoints not covered by dedicated tools.
Inputs: path (required), body, query, method, +1 more.
weflow_sns_block_deleteC
Manage WeFlow Moments anti-delete status. Actions map to status, install, and uninstall endpoints.
Inputs: action (required).
weflow_sns_exportC
Export WeFlow Moments to a local directory from the WeFlow app process.
Inputs: outputDir (required), end, start, format, +6 more.
weflow_sns_export_statsC
Get WeFlow Moments export statistics.
Inputs: fast.
weflow_sns_timelineC
Get WeFlow Moments timeline entries.
Inputs: end, limit, media, start, +5 more.
weflow_sns_usernamesB
List WeFlow Moments publishers.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 12 tools
The two message retrieval tools (weflow_get_messages and weflow_get_session_messages_chatlab) have overlapping purposes; the first already supports ChatLab format, making the second redundant and confusing. Other tools are distinct within their domains.
Tool names use a consistent 'weflow_' prefix but mix verb_noun (list_sessions, get_messages), noun-only (health, request), and SNS category prefixes (sns_timeline, sns_export) inconsistently. The use of both 'list_' and 'get_' for similar actions adds inconsistency.
12 tools is appropriate for a domain covering health, chat sessions, messages, contacts, group members, and Moments. The number is well within the expected range and each tool addresses a specific sub-area.
The server provides comprehensive read and export functionality for chat and Moments, but lacks dedicated tools for sending messages or managing contacts/groups. While weflow_request can fill some gaps, core operations like sending a message are not directly exposed.
Maintenance
Related MCP Connectors
AppFlowy MCP server for ChatGPT, Claude and any AI agent: list workspaces, read page tree, list database rows, create database row, update database row, create page, restore from trash and share page in your own AppFlowy account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to AppFlowy yourself; Cotlet does not store your password and you need no AppFlowy API key. About the official AppFlowy API: AppFlowy Cloud has a documented REST API for workspaces, the folder tree and database rows, but no hosted MCP server (checked 25 Oct 2026). The official MCP service is an optional part of the commercial self-hosted edition. Setup and FAQ: https://cotlet.dev/integrations/appflowy-api-mcp-server
WHOOP MCP server for ChatGPT, Claude and any AI agent: get day summary, list recoveries, list sleeps, list workouts, list cycles and get membership in your own WHOOP account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to WHOOP yourself; Cotlet does not store your password and you need no WHOOP API key. About the official WHOOP API: WHOOP has an official developer API (v2, OAuth 2.0, read-only) but no official MCP server (checked 27 Oct 2026). The API returns recovery, sleep, cycles and workouts to an app you register in the WHOOP Developer Dashboard, and an app WHOOP has not approved serves 10 members. Setup and FAQ: https://cotlet.dev/integrations/whoop-api-mcp-server
ManyChat MCP server for ChatGPT, Claude and any AI agent: list contacts, get contact profile, read inbox conversation, send inbox message, add tag to contact, set custom field, list broadcasts and send broadcast in your own ManyChat account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to ManyChat yourself; Cotlet does not store your password and you need no ManyChat API key. About the official ManyChat API: Yes, but only on the Pro plan or higher, and it is narrow (checked 5 Nov 2026). ManyChat has a REST API with a Page API and a Profile API that reads and sets contacts, tags and custom fields, sends content and triggers an automation for one contact. Setup and FAQ: https://cotlet.dev/integrations/manychat-api-mcp-server
Weezevent MCP server for ChatGPT, Claude and any AI agent: list events, list orders, list attendees, get participation report, create discount codes, send invitations, request refund and resend order in your own Weezevent account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to Weezevent yourself; Cotlet does not store your password and you need no Weezevent API key. About the official Weezevent API: Weezevent has a REST API, but it only reads tickets and attendees and it is documented for registered partners (checked 31 Oct 2026). It lists events, price categories, scan statistics and participants. Setup and FAQ: https://cotlet.dev/integrations/weezevent-api-mcp-server
Related MCP Servers
- FlicenseCqualityDmaintenanceMCP server for reading local WeChat data, enabling AI assistants to query chat history, contacts, sessions, and more via MCP tools.209-
- AlicenseBqualityBmaintenanceEnables MCP-compatible AI clients to read local WeChat data through an existing WxLens HTTP service, so users can find messages, extract a person's statements, summarize discussions, scan multiple groups, and read related articles or attachments.405MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to read local WeChat chat records, sessions, contacts, and group members through WeFlow's local HTTP API, keeping all data on localhost.MIT
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients on Windows to read WeChat sessions, contacts, messages, media, and files through WeFlow, and to send or natively forward messages via the desktop WeChat client.MIT