Omilia MCP Tools
A set of tools for working with Omilia Cloud Platform that helps manage miniapps, orchestrator apps, and dialog logs through an MCP server.
README
Omilia MCP Tools
This repository contains a set of tools for working with the Omilia Cloud Platform (OCP). These utilities help manage miniapps, orchestrator apps, and dialog logs.
Tools Overview
- search_miniapps: Search for miniapps by name or keyword.
- get_miniapp: Retrieve details for a specific miniapp using its ID.
- set_miniapp_prompt: Update prompts (welcome, error, reaction messages) for a miniapp.
- get_dialog_logs: Fetch logs for a specific dialog session.
- search_orchestrator_apps: Search for Orchestrator apps by keyword.
- get_orchestrator_app: Retrieve the canvas (nodes and edges) for an Orchestrator app by ID.
- search_dialog_logs: Search dialog logs with various filters (date, app, region, etc.).
- search_numbers: Search for phone numbers with optional search term.
- search_variable_collections: Search variable collections with optional search term.
- get_collection_variables: Get a list of all variables in a collection by ID.
Installation
- Make sure you have Python 3.10 or newer installed.
- Install uv.
- Clone this repository and navigate to the project directory.
- Copy the file
.env.exampleto.envand set the appropriate values. - Test if the istallation is correct by running
uv run mcp dev src/main.py. This should open the mcp development server. Click on connect and try it out.
Usage
You can use these tools in two main ways:
1. Self-hosting (MCP Python SDK)
You can run your own MCP server using the official Python MCP SDK. This is the most flexible option and is recommended for advanced users. For full instructions, see the MCP Python SDK README.
2. Local usage with Gemini CLI, Cursor, or Claude Desktop
You can also use this project locally with any MCP-compatible client, such as:
Each of these clients allows you to connect to local MCP servers. For more information, see their respective documentation:
- Gemini CLI: Configuring custom MCP servers
- Claude Desktop: MCP servers
- Cursor: Configuring custom MCP servers
Configuring MCP Servers
To use this project with any of the above clients, you need to configure your MCP servers. For example, you can use the following mcp.json configuration (place it in the appropriate config directory for your client):
{
"mcpServers": {
"Omilia MCP": {
"command": "uv",
"args": [
"run",
"--with",
"mcp",
"mcp",
"run",
"<path_to_cloned_repository>>/omilia-mcp/src/main.py"
],
"env": {
"PATH": "<depending on how you installed the needed tools you may need to paste your PATH here>"
}
}
}
}
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。