io.github.intenttext/intenttext-mcp

编码与调试

by intenttext

Parse, validate, query, and render IntentText (.it) documents for AI agents

什么是 io.github.intenttext/intenttext-mcp

Parse, validate, query, and render IntentText (.it) documents for AI agents

README

@intenttext/mcp-server

<!-- mcp-name: io.github.intenttext/intenttext-mcp -->

What is IntentText?

IntentText (.it) is a plain-text document format where every line has a declared semantic type — making documents simultaneously human-writable and natively JSON. A task: is a task. A step: is an executable workflow step. Every block parses to typed, deterministic JSON with no interpretation required.

This MCP server gives any AI agent the ability to work with IntentText documents as native tool calls.

Deployed instance: https://intenttext-mcp.onrender.com — use this URL directly in any MCP client that supports remote servers.


MCP server for IntentText — parse, validate, query, render, and generate .it documents from any AI agent.

With this server running, Claude, GPT, or any MCP-compatible agent can work with IntentText documents as native tool calls — without needing to understand the format itself.

Installation

bash
npm install -g @intenttext/mcp-server
# or use npx (no install required)
npx @intenttext/mcp-server

Configure with Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

json
{
  "mcpServers": {
    "intenttext": {
      "command": "npx",
      "args": ["@intenttext/mcp-server"]
    }
  }
}

Configure with any MCP client

json
{
  "command": "npx",
  "args": ["@intenttext/mcp-server"],
  "env": {}
}

HTTP Wrapper (for hosted MCP URLs)

This package also includes an HTTP wrapper so you can deploy it from this GitHub repo to platforms like Railway/Render/Fly and provide a URL (for directories that ask for https://.../mcp).

bash
npm install
npm run build
npm run start:http

Endpoints:

  • POST /mcp - MCP Streamable HTTP endpoint
  • GET /health - health check endpoint

Environment variables:

  • PORT (default 3000)
  • HOST (default 0.0.0.0)

Note: GitHub itself cannot host a long-running Node server process. Keep the wrapper code in GitHub, then deploy it to a runtime provider and use that public URL in Smithery forms.

The public deployment is available at https://intenttext-mcp.onrender.com.

Available Tools

parse_intent_text

Parse an IntentText (.it) source string into a structured JSON document.

ParameterTypeDescription
sourcestringThe IntentText source string to parse
safebooleanIf true, never throw — returns warnings instead of errors. Default: true

render_html

Render an IntentText source string to styled HTML.

ParameterTypeDescription
sourcestringIntentText source string (.it format)

render_print

Render an IntentText document to print-optimised HTML with @media print CSS. Suitable for PDF generation.

ParameterTypeDescription
sourcestringIntentText source string

merge_template

Merge an IntentText template (containing {{variable}} placeholders) with a data object.

ParameterTypeDescription
templatestringIntentText template source with {{variable}} placeholders
dataobjectJSON object with values to substitute into the template
renderstring"none" (default), "html", or "print" — optionally render the merged result

validate_document

Validate an IntentText document for semantic correctness. Checks for broken step references, missing required properties, unresolved variables, and workflow logic errors.

ParameterTypeDescription
sourcestringIntentText source string to validate

query_document

Query an IntentText document for specific blocks by type, content, or properties.

ParameterTypeDescription
sourcestringIntentText source string
typestringBlock type filter — e.g. "task", "step,gate,decision"
contentstringSubstring to search for in block content
sectionstringOnly return blocks within this section
limitnumberMaximum number of results to return

diff_documents

Compare two versions of an IntentText document and return a semantic diff.

ParameterTypeDescription
beforestringThe original IntentText source
afterstringThe updated IntentText source

document_to_source

Convert an IntentText JSON document back to .it source format.

ParameterTypeDescription
documentobjectAn IntentText document JSON object (as produced by parse_intent_text)

extract_workflow

Extract the execution graph from an IntentText workflow document. Returns steps in topological order, dependency relationships, parallel batches, and gate positions.

ParameterTypeDescription
sourcestringIntentText source containing workflow blocks (step:, decision:, gate:, etc.)

Example: Agent Workflow

code
User: Create a deployment workflow for a web app

Agent calls: parse_intent_text with a generated .it document
Agent calls: validate_document to check for issues
Agent calls: extract_workflow to get the execution graph
Agent calls: render_html to produce a visual version

License

MIT

常见问题

io.github.intenttext/intenttext-mcp 是什么?

Parse, validate, query, and render IntentText (.it) documents for AI agents

相关 Skills

网页构建器

by anthropics

Universal
热门

面向复杂 claude.ai HTML artifact 开发,快速初始化 React + Tailwind CSS + shadcn/ui 项目并打包为单文件 HTML,适合需要状态管理、路由或多组件交互的页面。

在 claude.ai 里做复杂网页 Artifact 很省心,多组件、状态和路由都能顺手搭起来,React、Tailwind 与 shadcn/ui 组合效率高、成品也更精致。

编码与调试
未扫描114.1k

前端设计

by anthropics

Universal
热门

面向组件、页面、海报和 Web 应用开发,按鲜明视觉方向生成可直接落地的前端代码与高质感 UI,适合做 landing page、Dashboard 或美化现有界面,避开千篇一律的 AI 审美。

想把页面做得既能上线又有设计感,就用前端设计:组件到整站都能产出,难得的是能避开千篇一律的 AI 味。

编码与调试
未扫描114.1k

网页应用测试

by anthropics

Universal
热门

用 Playwright 为本地 Web 应用编写自动化测试,支持启动开发服务器、校验前端交互、排查 UI 异常、抓取截图与浏览器日志,适合调试动态页面和回归验证。

借助 Playwright 一站式验证本地 Web 应用前端功能,调 UI 时还能同步查看日志和截图,定位问题更快。

编码与调试
未扫描114.1k

相关 MCP Server

GitHub

编辑精选

by GitHub

热门

GitHub 是 MCP 官方参考服务器,让 Claude 直接读写你的代码仓库和 Issues。

这个参考服务器解决了开发者想让 AI 安全访问 GitHub 数据的问题,适合需要自动化代码审查或 Issue 管理的团队。但注意它只是参考实现,生产环境得自己加固安全。

编码与调试
83.4k

by Context7

热门

Context7 是实时拉取最新文档和代码示例的智能助手,让你告别过时资料。

它能解决开发者查找文档时信息滞后的问题,特别适合快速上手新库或跟进更新。不过,依赖外部源可能导致偶尔的数据延迟,建议结合官方文档使用。

编码与调试
52.2k

by tldraw

热门

tldraw 是让 AI 助手直接在无限画布上绘图和协作的 MCP 服务器。

这解决了 AI 只能输出文本、无法视觉化协作的痛点——想象让 Claude 帮你画流程图或白板讨论。最适合需要快速原型设计或头脑风暴的开发者。不过,目前它只是个基础连接器,你得自己搭建画布应用才能发挥全部潜力。

编码与调试
46.3k

评论