Pararam Nexus MCP
Enables interaction with pararam.io communication platform, allowing users to search and send messages, manage chats, upload/download files, and search users through natural language commands.
README
Pararam Nexus MCP
A Model Context Protocol (MCP) server for interacting with pararam.io - a modern communication and collaboration platform.
About pararam.io
Pararam.io is a communication platform that provides:
- Messaging: Create groups and private chats
- Team Organization: Organize people into teams
- Communication: Group and private calls (audio and video)
- Cross-platform: Available on mobile (iOS, Android, Huawei) and web
This MCP server uses the pararamio-aio library to provide asynchronous access to pararam.io features through the Model Context Protocol.
Features
- Asynchronous API client for pararam.io
- Two-factor authentication support (TOTP)
- Session persistence with cookie storage
- Comprehensive chat and message management
- File attachment handling (upload and download)
- URL-based message retrieval
- Conversation thread building
Available Tools
Message Operations
- search_messages: Search for messages across all chats with advanced search syntax (Boolean operators, wildcards, filters)
- get_chat_messages: Get recent messages from a specific chat
- send_message: Send a message to a chat with optional reply and quote text
- get_message_from_url: Extract and retrieve a message from pararam.io URL
Chat Operations
- search_chats: Search for chats by name or description
- build_conversation_thread: Build a conversation tree from a root message
File Operations
- upload_file_to_chat: Upload files to a chat (from path or base64 content)
- get_post_attachments: List all attachments in a post
- download_post_attachment: Download attachments (to disk or as ImageContent)
- 1MB size limit for downloads
- Supported formats for direct display: images (JPEG, PNG, GIF, WEBP), documents (PDF, DOCX, DOC, TXT, RTF, ODT, HTML, EPUB), spreadsheets (XLSX, XLS, CSV), data (JSON, XML)
- Returns ImageContent for supported types (displays natively in Claude Desktop/Code)
- For unsupported types, requires output_path to save to disk
- Saves to disk when output path is provided
User Operations
- search_users: Search for users by name or unique name
- get_user_info: Get detailed information about a specific user
- get_user_team_status: Get user's status in teams (member, admin, guest)
Tool Details
send_message
Send a message to a chat with optional reply and quote functionality.
Parameters:
chat_id(required): ID of the chat to send message totext(required): Message text to sendreply_to_message_id(optional): Post number to reply toquote_text(optional): Text to quote from the replied message (only used withreply_to_message_id)
Examples:
# Simple message
send_message(chat_id="123", text="Hello!")
# Reply to a message
send_message(
chat_id="123",
text="I agree!",
reply_to_message_id="456"
)
# Reply with quoted text
send_message(
chat_id="123",
text="That's a great idea!",
reply_to_message_id="456",
quote_text="We should implement this feature next week"
)
Installation
Quick Install with uvx (Recommended)
Install directly from PyPI:
uvx pararam-nexus-mcp
Or install from GitHub:
uvx --from git+https://github.com/ivolnistov/pararam-nexus-mcp pararam-nexus-mcp
Or clone to a specific directory (e.g., ~/.mcp/):
git clone https://github.com/ivolnistov/pararam-nexus-mcp.git ~/.mcp/pararam-nexus-mcp
cd ~/.mcp/pararam-nexus-mcp
uv sync
Docker Installation
Pull from Docker Hub:
docker pull ivolnistov/pararam-nexus-mcp:latest
Or from GitHub Container Registry:
docker pull ghcr.io/ivolnistov/pararam-nexus-mcp:latest
Development Installation
For local development:
git clone https://github.com/ivolnistov/pararam-nexus-mcp.git
cd pararam-nexus-mcp
uv sync --dev
Configuration
Create a .env file with your pararam.io credentials:
PARARAM_LOGIN=your_login
PARARAM_PASSWORD=your_password
PARARAM_2FA_KEY=your_2fa_key # Optional
MCP Client Configuration
Claude Code (CLI)
Add the server using the Claude Code CLI:
# Using uvx (recommended)
claude mcp add pararam-nexus \
--env PARARAM_LOGIN=myuser@example.com \
--env PARARAM_PASSWORD=mySecurePassword123 \
--env PARARAM_2FA_KEY=JBSWY3DPEHPK3PXP \
-- uvx pararam-nexus-mcp
# Using Docker
claude mcp add pararam-nexus \
--env PARARAM_LOGIN=myuser@example.com \
--env PARARAM_PASSWORD=mySecurePassword123 \
--env PARARAM_2FA_KEY=JBSWY3DPEHPK3PXP \
-- docker run -i --rm ivolnistov/pararam-nexus-mcp:latest
Claude Desktop
Option 1: Using uvx (Recommended)
- Open Claude Desktop preferences
- Navigate to the MCP section
- Click Edit to open
claude_desktop_config.json - Add the server configuration:
{
"mcpServers": {
"pararam-nexus": {
"command": "uvx",
"args": ["pararam-nexus-mcp"],
"env": {
"PARARAM_LOGIN": "myuser@example.com",
"PARARAM_PASSWORD": "mySecurePassword123",
"PARARAM_2FA_KEY": "JBSWY3DPEHPK3PXP"
}
}
}
}
Option 2: Using Docker
First, create a volume for session persistence:
docker volume create pararam-mcp-data
Then add to claude_desktop_config.json:
{
"mcpServers": {
"pararam-nexus": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-v",
"pararam-mcp-data:/app/.cookies",
"-e",
"PARARAM_LOGIN=myuser@example.com",
"-e",
"PARARAM_PASSWORD=mySecurePassword123",
"-e",
"PARARAM_2FA_KEY=JBSWY3DPEHPK3PXP",
"ivolnistov/pararam-nexus-mcp:latest"
]
}
}
}
Cursor IDE
Add to .cursor/mcp.json:
{
"mcpServers": {
"pararam-nexus": {
"command": "uvx",
"args": ["pararam-nexus-mcp"],
"env": {
"PARARAM_LOGIN": "myuser@example.com",
"PARARAM_PASSWORD": "mySecurePassword123",
"PARARAM_2FA_KEY": "JBSWY3DPEHPK3PXP"
}
}
}
}
Or using Docker (with session persistence):
First, create a volume:
docker volume create pararam-mcp-data
Then add to .cursor/mcp.json:
{
"mcpServers": {
"pararam-nexus": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-v",
"pararam-mcp-data:/app/.cookies",
"-e",
"PARARAM_LOGIN=myuser@example.com",
"-e",
"PARARAM_PASSWORD=mySecurePassword123",
"-e",
"PARARAM_2FA_KEY=JBSWY3DPEHPK3PXP",
"ivolnistov/pararam-nexus-mcp:latest"
]
}
}
}
Other MCP-Compatible Clients
For any MCP-compatible client that supports stdio transport:
Using uvx:
uvx pararam-nexus-mcp
Using Docker (with session persistence):
# Create volume for cookies
docker volume create pararam-mcp-data
# Run with volume mounted
docker run -i --rm \
-v pararam-mcp-data:/app/.cookies \
-e PARARAM_LOGIN=myuser@example.com \
-e PARARAM_PASSWORD=mySecurePassword123 \
-e PARARAM_2FA_KEY=JBSWY3DPEHPK3PXP \
ivolnistov/pararam-nexus-mcp:latest
Environment variables:
PARARAM_LOGIN(required): Your pararam.io loginPARARAM_PASSWORD(required): Your pararam.io passwordPARARAM_2FA_KEY(optional): Your 2FA secret key for TOTP authentication
Usage
If installed with uvx:
uvx --from git+https://github.com/ivolnistov/pararam-nexus-mcp pararam-nexus-mcp
If cloned locally:
cd ~/.mcp/pararam-nexus-mcp
uv run pararam-nexus-mcp
For development:
uv run pararam-nexus-mcp
Development
Install pre-commit hooks:
uv run pre-commit install
Run linting and formatting:
uv run ruff check --fix src/
uv run ruff format src/
Run type checking:
uv run mypy src/pararam_nexus_mcp
Run tests:
uv run pytest
Dependencies
- FastMCP: Model Context Protocol server framework
- pararamio-aio: Async Python client for pararam.io API
- httpx: Modern HTTP client
- Pydantic: Data validation using Python type annotations
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。