Intentions MCP Server
Provides AI agents with personalized timing intelligence by scoring decisions (0-100) against a user's energy profile and the Five Elements framework, enabling optimal scheduling for actions like trip planning, product launches, and meetings.
README
<p align="center"> <img src="./logo.png" width="120" height="120" alt="Intentions" /> </p>
<h1 align="center">Intentions MCP Server</h1>
Personalized timing intelligence for AI agents. Ask "Should I do X on this date?" and get a 0–100 alignment score, a clear verdict, and the reasoning behind it — grounded in the user's personal energy profile and the Five Elements framework.
Built for agentic workflows: trip planning, product launches, content calendars, interview/meeting scheduling, negotiation prep, and relationship/family timing.
- Homepage: https://www.intentions.me
- MCP endpoint:
https://mcp.intentions.me/mcp(Streamable HTTP)
What it does
You give it a decision and a time window; it returns a structured reading your agent can reason over and explain in plain language:
- Score (0–100) and a verdict (e.g. favorable, proceed with caution, wait)
- Friction (0–100) — an independent measure of execution resistance
- Element breakdown and per-layer (year / month / day / hour) signals
- Adverse alerts with actionable, behavioral guidance — never event predictions
Chain the tools to narrow from "which year" → "which month" → "which day" → "which hour".
Tools
| Tool | What it answers | Tier |
|---|---|---|
intentions_ask_year |
A full calendar year as a whole | Free |
intentions_ask_month |
A specific month, or the best month in a window (≤12 months) | Free |
intentions_ask_day |
A specific date, comparison, or best day in a range (≤31 days) | Free |
intentions_ask_hour |
Best hour within a day; morning vs afternoon | Pro |
intentions_energy_chart |
ASCII chart overview — weekly / yearly (free), hourly (Pro) |
Mixed |
intentions_set_birth_info |
Register or update the user's birth date / time / gender | Free |
intentions_get_profile |
Read the currently stored birth info | Free |
All timing inputs must fall within currentYear − 1 … currentYear + 1.
Installation
The server is remote (Streamable HTTP) — nothing to install or self-host.
- Create an account at https://www.intentions.me
- Generate an API key from your account page (format:
itm_…) - Add the server to your MCP client using the config below
Claude Desktop / MCP client config
{
"mcpServers": {
"intentions": {
"type": "http",
"url": "https://mcp.intentions.me/mcp",
"headers": {
"Authorization": "Bearer itm_your_api_key_here"
}
}
}
}
claude.ai custom connector
Add a custom connector with:
- URL:
https://mcp.intentions.me/mcp - Authentication: Bearer token — paste your
itm_…key
Authentication
Every request authenticates with a Bearer API key in the Authorization header:
Authorization: Bearer itm_0123456789abcdef0123456789abcdef
Keys are issued per account at https://www.intentions.me/account. Treat the key like a password — it carries your quota and your stored energy profile.
Pricing
| Free | Pro | |
|---|---|---|
| Day / month / year readings | 5 per day (resets midnight UTC) | Unlimited |
Hour-by-hour precision (ask_hour, hourly charts) |
— | ✓ |
| Weekly / yearly charts | ✓ | ✓ |
The free tier is enough to evaluate the server end-to-end. See https://www.intentions.me/pricing for current Pro pricing.
Links
- Homepage: https://www.intentions.me
- Pricing: https://www.intentions.me/pricing
This repository contains documentation only. The Intentions MCP Server is a hosted, proprietary service; the implementation source is not open source.
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。