About Mcp Server For Bitrix24
Mcp Server For Bitrix24 connects MCP clients to Bitrix24 Tasks, Workgroups, and Users through a webhook. It is intended for workflows that manage tasks and inspect organization data, with task-field validation and history access.
Key features
- 01Create, update, close, and reopen tasks
- 02List tasks and task history
- 03Create and list workgroups
- 04List users and inspect user fields
- 05Validate task fields against Bitrix24 fields
Use cases
- 01Manage task status and assignments
- 02Review task changes over time
- 03Create workgroups for project organization
MCP support
Exposes Bitrix24 task, workgroup, and user operations as MCP tools. It uses MCP/JSON-RPC over STDIO and requires a Bitrix24 webhook URL.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
12Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| title | string | Yes | |
| fields | object | No | |
| group_id | number | No | |
| description | string | No | |
| responsible_id | number | No |
Details
From the project READMEmcp-bitrix24
MCP server for Bitrix24 Tasks, Workgroups, and Users. Implements MCP/JSON-RPC over STDIO.
Features
- Tasks: create, update, close, reopen, list
- Workgroups: create, list
- Users: list, current user, available fields
- Task fields: available fields + validation for
create_task.fields
Requirements
- Node.js >= 18
- Bitrix24 webhook URL
Install / Build
npm install
npm run build
Run via npm:
npx mcp-bitrix24
Configuration
Set the Bitrix24 webhook URL via environment variable:
BITRIX24_WEBHOOK_URL=https://<your-domain>/rest/<user_id>/<webhook>/
Example Codex MCP config:
[mcp_servers.bitrix24]
command = "npx"
args = ["-y", "mcp-bitrix24"]
[mcp_servers.bitrix24.env]
BITRIX24_WEBHOOK_URL = "https://<your-domain>/rest/<user_id>/<webhook>/"
Tools
Tasks
-
create_task- Input:
title(string, required),description?(string),responsible_id?(number),group_id?(number),fields?(object) - Output:
{ task_id: number } - Note: if
fieldsis provided, keys are validated againstget_task_fields.
- Input:
-
update_task- Input:
task_id(number, required) + at least one of:title?,description?,responsible_id?,group_id? - Output:
{ task_id: number }
- Input:
-
close_task- Input:
task_id(number, required) - Output:
{ task_id: number }
- Input:
-
reopen_task- Input:
task_id(number, required) - Output:
{ task_id: number }
- Input:
-
list_tasks- Input:
responsible_id?(number),group_id?(number),start?(number),limit?(number) - Output:
{ tasks: [{ id, title, status }] }
- Input:
-
get_task_fields- Input:
{} - Output:
{ fields: { [field: string]: object } }
- Input:
-
list_task_history- Input:
task_id(number, required),filter?(object),order?(object) - Output:
{ list: [ { id, createdDate, field, value, user } ] }
- Input:
Workgroups
-
create_group- Input:
name(string, required),description?(string) - Output:
{ group_id: number }
- Input:
-
list_groups- Input:
limit?(number) - Output:
{ groups: [{ id, name }] }
- Input:
Users
-
list_users- Input:
filter?(object)sort?(string)order?("ASC" | "DESC")admin_mode?(boolean)start?(number)limit?(number)
- Output:
{ users: [{ id, name, last_name, email?, active }] } - Note:
filtersupports Bitrix24user.getfilters (including prefixes like>=,%,@, etc.).startcontrols paging (Bitrix returns 50 records per page);limitis a local slice after the API response.
- Input:
-
get_user_fields- Input:
{} - Output:
{ fields: { [field: string]: string } }
- Input:
-
get_current_user- Input:
{} - Output:
{ user: { id, name, last_name, email?, active } }
- Input:
Architecture
Clean architecture layers:
mcp/— protocol, transport, serveradapters/— MCP tools mapping to domaindomain/— entities, services, portsinfrastructure/— Bitrix24 REST client
Development Notes
- Input validation uses
zod. - Transport: STDIO only.
- Build:
tsc(npm run build).
Contributing
See CONTRIBUTING.md for guidelines.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.