Planfix MCP Server
Enables integration with Planfix CRM and task management system, allowing lead management, contact/company management, task creation/search/commenting, and report generation via the Planfix REST API v2.0.
README
Planfix MCP Server
This MCP server provides integration with the Planfix API, allowing Model Context Protocol (MCP) clients to interact with Planfix CRM and task management system.
Features
- Lead management (create, search, convert to tasks)
- Lead searches can reuse a known
clientIdto skip contact lookups - Contact and company management
- Task management (create, search, comment)
- Report generation and management
- Uses Planfix REST API v2.0 (API docs)
- Authentication via Bearer token
Configuration
The server requires the following environment variables for Planfix API access:
PLANFIX_ACCOUNT– Your Planfix account name (e.g.,yourcompany)PLANFIX_TOKEN– Planfix API token with necessary permissionsPLANFIX_BASE_URL– (optional) Override the REST API base URL. Defaults tohttps://<PLANFIX_ACCOUNT>.planfix.com/rest/. Set this for.ruand other regional installations, e.g.https://yourcompany.planfix.ru/rest/PLANFIX_ACCOUNT_URL– (optional) Override the web origin used for human-facing links (task/contact/user pages). Defaults toPLANFIX_BASE_URLwithout the trailing/rest/PLANFIX_FIELD_ID_EMAIL– Custom field ID for emailPLANFIX_FIELD_ID_PHONE– Custom field ID for phonePLANFIX_FIELD_ID_TELEGRAM– Set any value to use the system Telegram fieldPLANFIX_FIELD_ID_TELEGRAM_CUSTOM– Custom field ID for Telegram when using the custom fieldPLANFIX_FIELD_ID_CLIENT– Custom field ID for clientPLANFIX_FIELD_ID_MANAGER– Custom field ID for managerPLANFIX_FIELD_ID_AGENCY– Custom field ID for agencyPLANFIX_FIELD_ID_LEAD_SOURCE– Custom field ID for lead sourcePLANFIX_FIELD_ID_LEAD_SOURCE_VALUE– Value ID for default lead sourcePLANFIX_FIELD_ID_PIPELINE– Custom field ID for pipelinePLANFIX_FIELD_ID_TAGS– Custom field ID for task tags- Missing tag names will be added automatically to the directory
PLANFIX_FIELD_ID_LEAD_ID– Custom field ID for external lead IDPLANFIX_LEAD_TEMPLATE_ID– ID of the lead task templatePLANFIX_TASK_TITLE_TEMPLATE– Template for the default lead task title (e.g.,{name} - client's task)
config.yml
Custom fields can also be configured via config.yml. The default path is
./data/config.yml. Override it with the --config=/abs/path/config.yml CLI
flag or the PLANFIX_CONFIG environment variable. You can also specify a
different Planfix account when using a custom config:
PLANFIX_CONFIG=/etc/planfix-mcp.yml PLANFIX_ACCOUNT=demo \
npx @popstas/planfix-mcp-server
proxyUrl: "http://localhost:8080"
webhook:
enabled: false
url: "https://example.com/hook"
token: "<token>"
skipPlanfixApi: false
leadTaskFields:
- id: "456"
name: "id сделки"
argName: lead_id
type: number
contactFields:
- id: "123"
name: "Резидентство"
argName: resident
type: enum
values: ["резидент", "нерезидент", "иное"]
userFields:
- id: "789"
name: "Департамент"
argName: department
type: string
proxyUrl routes all Planfix REST API calls (including tool requests) through
the specified HTTP proxy.
Values from config.yml override matching entries from the legacy environment
variables when merged by id. User custom fields from this list are requested
individually by the planfix_search_manager tool so their values are available
in responses. Managers can be searched either by email or by numeric id
through this tool, enabling lookups when only an identifier is available.
Chat API
To create tasks from chat messages, add a chatApi block to config.yml:
chatApi:
useChatApi: true
chatApiToken: "<token>"
providerId: "<id>"
baseUrl: "https://<account>.planfix.com/webchat/api"
chatApiToken– token for Planfix Chat API requests.providerId– identifier of the chat provider configured in Planfix.useChatApi– enable Chat API integration. Whentrue, task creation proceeds as:- A chat is created via Chat API with the initial message.
getTaskretrieves the new task'staskId.- Subsequent updates are made through the REST API.
baseUrl– base URL for Chat API calls. Defaults tohttps://<account>.planfix.com/webchat/api.
Webhook
To post lead task payloads to a webhook before creating or updating a task,
add a webhook block to config.yml:
webhook:
enabled: true
url: "https://example.com/hook"
token: "<token>"
skipPlanfixApi: false
enabled– whether to send the lead task payload to the webhook URL.url– webhook endpoint URL.token– shared secret appended to the JSON payload astoken.skipPlanfixApi– whentrue, the webhook response must includetaskId, and the Planfix REST API call is skipped.
Debug
npx @modelcontextprotocol/inspector node d:/projects/expertizeme/planfix-mcp-server/dist/index.js
Logging
Set LOG_LEVEL=debug to enable detailed cache logs. Logs are written to data/mcp.log.
Clearing Cache
Run npm run cache-clear to remove all cached Planfix API responses stored in data/planfix-cache.sqlite3 and delete the objects cache file data/planfix-cache.yml.
Example MCP Config (NPX)
{
"mcpServers": {
"planfix": {
"command": "npx",
"args": [
"-y",
"@popstas/planfix-mcp-server"
],
"env": {
"PLANFIX_ACCOUNT": "yourcompany",
"PLANFIX_TOKEN": "your-api-token",
"PLANFIX_FIELD_ID_EMAIL": "123",
"PLANFIX_FIELD_ID_PHONE": "124",
"PLANFIX_FIELD_ID_TELEGRAM": "1",
"PLANFIX_FIELD_ID_TELEGRAM_CUSTOM": "125",
"PLANFIX_FIELD_ID_CLIENT": "126",
"PLANFIX_FIELD_ID_MANAGER": "127",
"PLANFIX_FIELD_ID_AGENCY": "128",
"PLANFIX_FIELD_ID_TAGS": "129",
"PLANFIX_FIELD_ID_LEAD_ID": "130",
"PLANFIX_LEAD_TEMPLATE_ID": "42",
"PLANFIX_TASK_TITLE_TEMPLATE": "{name} - работа с клиентом"
}
}
}
}
Usage
Running the Server
Run the server with the required environment variables set. Example (with npx):
PLANFIX_ACCOUNT=yourcompany \
PLANFIX_TOKEN=your-api-token \
PLANFIX_FIELD_ID_EMAIL=123 \
PLANFIX_FIELD_ID_PHONE=124 \
PLANFIX_FIELD_ID_TELEGRAM=1 \
PLANFIX_FIELD_ID_TELEGRAM_CUSTOM=125 \
PLANFIX_FIELD_ID_CLIENT=126 \
PLANFIX_FIELD_ID_MANAGER=127 \
PLANFIX_FIELD_ID_AGENCY=128 \
PLANFIX_FIELD_ID_LEAD_SOURCE=129 \
PLANFIX_FIELD_ID_LEAD_SOURCE_VALUE=130 \
PLANFIX_FIELD_ID_PIPELINE=131 \
PLANFIX_FIELD_ID_LEAD_ID=132 \
PLANFIX_FIELD_ID_TAGS=133 \
PLANFIX_LEAD_TEMPLATE_ID=42 \
PLANFIX_TASK_TITLE_TEMPLATE="{name} - работа с клиентом" \
npx @popstas/planfix-mcp-server
To run the server over Server-Sent Events (SSE), use the planfix-mcp-server-sse command:
PLANFIX_ACCOUNT=yourcompany \
PLANFIX_TOKEN=your-api-token \
planfix-mcp-server-sse
Using the Planfix Client
The Planfix client provides a convenient way to interact with the Planfix API directly from the command line.
Prerequisites
Make sure you have the following environment variables set in your .env file:
PLANFIX_ACCOUNT=your-account
PLANFIX_TOKEN=your-api-token
Basic Commands
-
Test the connection
npm run planfix test -
Make a GET request
npm run planfix get user/current -
Make a POST request with data
npm run planfix post task/ --data '{"name":"Test Task","description":"Test Description"}' -
Search for objects
npm run planfix post object/list --data '{"filters":[{"type":1,"operator":"equal","value":"Продажа"}]}'
Tool Reference
planfix_create_sell_task
- Creates a sell task using textual information about the agency and employee.
- Resolves the client, parent lead task, assignees, and agency IDs automatically based on the provided strings.
- Input fields (all strings):
name: Task title, e.g."Продажа {{ название товара }} на pressfinity.com".agency: Agency/company name (optional).email: Employee email used to locate the Planfix contact.contactName/employeeName: Employee full name (optional).telegram: Employee telegram username (optional).description: Description with the list of ordered products.project: Project name to associate with the sell task (optional).
- Returns
{ taskId, url }.
planfix_create_sell_task_ids
- Creates a sell task when Planfix identifiers are already known.
- Requires numeric
clientIdand optionalleadTaskId,agencyId, andassignees(user IDs). - Accepts
name,description, and optionalprojectstring values.
-
Update an object (PUT request)
npm run planfix put task/123 --data '{"name":"Updated Task Name"}' -
Delete an object
npm run planfix delete task/123
Using in Code
import { planfixClient } from './lib/planfix-client';
// Get current user
const user = await planfixClient.get('user/current');
// Create a new task
const newTask = await planfixClient.post('task/', {
name: 'New Task',
description: 'Task description',
// ... other task properties
});
// Search for objects
const objects = await planfixClient.post('object/list', {
filters: [
{
type: 1,
operator: 'equal',
value: 'Продажа'
}
]
});
Available Tools
Lead Management
leadToTask: Convert a lead to a task by creating/updating contact and tasksearchLeadTask: Search for lead tasks by contact information
Contact Management
searchPlanfixContact: Search contacts by name, phone, email, or TelegramcreatePlanfixContact: Create a new contact in PlanfixupdatePlanfixContact: Update existing contact informationsearchPlanfixCompany: Search for companies by name
Task Management
searchPlanfixTask: Search for tasks by title, client ID and optionaltemplateIdcreateSellTask: Resolve contact/agency IDs and create a sell taskcreateSellTaskIds: Create a sell task when IDs are already knowncreateLeadTask: Create a new lead task. WhenchatApi.useChatApiis enabled, it sends the initial message through the Chat API, gets the resultingtaskIdviagetTask, and then updates the task using the REST API. AcceptsmessageandcontactNamefields.addToLeadTask: Create or update a lead task and update contact details. Whenwebhook.enabledis true, it posts the input payload to the webhook endpoint, optionally skipping the Planfix API ifskipPlanfixApiis set.createTask: Create a task using text fieldscreateComment: Add a comment to a taskgetChildTasks: Retrieve child tasks of a parent task. Userecursiveto fetch all descendant tasks as a flat list; returned tasks includeparent_task_id.updateLeadTask: Update an existing lead task (only empty fields are updated unlessforceUpdateis true)
Directory Management
planfix_search_directory: Search directories by nameplanfix_search_directory_entry: Search directory entry by directory name and entry name
User Management
searchManager: Find a manager by email
Reporting
listReports: List all available reportsrunReport: Generate and retrieve a specific report
References
TODO:
- Add tool
getTaskto retrieve task details - Add tool
getContactto retrieve contact details - Add tool
getManagerto retrieve manager details - Add more comprehensive error handling and logging
- Add input validation for all API endpoints
- Add rate limiting and retry logic for API calls
MIT License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.