Offorte Proposal Software
Offorte Proposal Software
README
Offorte MCP Server <!-- omit in toc -->
MCP server for Offorte - Create & send proposals using AI.
This server acts as the bridge between AI agents and Offorte's proposal engine. It enables external models to create and send proposals via Offorte. Built for automation workflows, the MCP makes it easy to integrate proposal actions into AI tools, chat interfaces, and autonomous systems.
Early technology Please note that MCP (Model Context Protocol) is a new approach to AI integration. While powerful, it's still evolving and may occasionally produce unexpected or undesired results.
Table of Contents <!-- omit in toc -->
About Offorte
Offorte is automated proposal software which helps businesses create, send, and track beautiful interactive proposals. Built for speed, flexibility, and real-world use, it combines automation and smart workflows without sacrificing the personal touch. Learn more about Offorte.
API <!-- omit in toc -->
The MCP server is using the Offorte Public API v2. Read the API documentation for more information.
Demo <!-- omit in toc -->
Experience the future of proposals: voice-triggered, AI-powered, fully automated. See how Offorte connects voice and workflow in this real demo.
Goals & Coverage
The goal of this project is too create & send proposals using AI via the protocol. Because of the experimental character, the full Offorte API is not covered and its limited to the tools needed to create & send proposals. Tools which could lead to an LLM updating and deleting stuff are not yet implemented. Update and delete commands might be added in the future, based on reallife results & user feedback.
Prerequisites
- Node.js (tested with Node.js
20.x.x) - Offorte API Key (see Authentication Section of the Offorte API Docs)
- PNPM for development
TRANSPORT_TYPE(optional): Set tosseto enable Server-Sent Events (SSE) mode, or leave unset/default forstdio(default:stdio).
Available Tools
Context & Setup <!-- omit in toc -->
- get_initial_context – IMPORTANT: Must be called before using any other tools to initialize context and get usage instructions
Account <!-- omit in toc -->
- get_users – Lists all account users for the current account
Automations <!-- omit in toc -->
- get_automation_sets – Lists automation sets which are used as an optional input to create a new proposal
Contacts <!-- omit in toc -->
- create_contact – Create a new contact (organisation or person/individual)
- get_contact_details - Get all details for a contact by id
- search_contact_organisations - Search for organisations by name in the contacts
- search_contact_people - Search for people by name in the contacts
Favorites <!-- omit in toc -->
- get_proposal_templates – Lists proposal templates which are used as starting points to create new proposals
Proposals <!-- omit in toc -->
- create_proposal – Create a new proposal
- get_proposal_directories – Get all proposal directories grouped by status
- search_proposals – Search for proposals by query
- send_proposal – Send a proposal to its assigned contacts
Settings <!-- omit in toc -->
- get_design_templates – Lists available design templates which are used to create new proposals
- get_email_templates – Lists available email templates which are used to send proposals
- get_text_templates – Lists available language text templates which are used to create new proposals
MCP Clients
Currently, this MCP server has only been tested with Claude Desktop. More client examples will be added in the future.
Claude Desktop Configuration <!-- omit in toc -->
Find your claude_desktop_config.json at Claude > Settings > Developer > Edit Config and depending on which option you'd like, add JUST ONE of the following:
NPX <!-- omit in toc -->
Running it straight from the npm registry.
{
"mcpServers": {
"offorte-proposals": {
"command": "npx",
"args": ["-y", "@offorte/mcp-server"],
"env": {
"OFFORTE_ACCOUNT_NAME": "<YOUR_ACCOUNT_NAME>",
"OFFORTE_API_KEY": "<YOUR_TOKEN>"
}
}
}
}
Local Node <!-- omit in toc -->
Dependencies should have been installed & the project is build before you use this method (pnpm install).
{
"mcpServers": {
"offorte-proposals": {
"command": "node",
"args": ["/path/to/directory/offorte-mcp-server/dist/server.js"],
"env": {
"OFFORTE_ACCOUNT_NAME": "<YOUR_ACCOUNT_NAME>",
"OFFORTE_API_KEY": "<YOUR_TOKEN>"
}
}
}
}
Development
To get started, clone the repository and install the dependencies. Make sure you have an .env file and it includes the proper values.
git clone https://github.com/offorte/offorte-mcp-server.git
cd offorte-mcp-server
pnpm install
pnpm dev
Commands <!-- omit in toc -->
Check the NPM scripts for all commands, below is a short summary of the most important onces.
| Script | Info |
|---|---|
pnpm build |
Build the project for production |
pnpm start |
Start the production server |
pnpm dev |
Start the development server |
pnpm inspect |
Inspect the server |
pnpm format |
Format code using Prettier and ESLint |
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。