Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mofox-docs-mcp how do I deploy Neo-MoFox on Windows?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Neo-MoFox official documentation MCP server: provides search and read capabilities for the MoFox-Bot-Docs documentation site via Model Context Protocol, so MCP clients such as Claude, Cursor, and Neo-MoFox can directly search and read the official documentation.
The data source is the documentation site's JSON API (https://docs.mofox-sama.com/api/docs/*), which shares the same llms.json index as the mofox_docs plugin in the MoFox plugin marketplace, so the behavior is consistent.
Tool | Description |
| Searches documents by keyword. The query is split by whitespace, and results are returned in descending order of the number of matched keywords (for example, |
| Lists documents. Sorted by section, with deployment guides ( |
| Gets the full body of the specified document (plain text). |
Parameter | Description |
| Full index ( |
| Full text of a single document (e.g. |
Related MCP server: mdbook-mcp-server
npx -y mofox-docs-mcpnpm install
npm run build
npm startEdit config/mcp.toml:
[mcp.stdio_servers.mofox-docs]
command = "npx"
args = ["-y", "mofox-docs-mcp"]
instructions = "提供 Neo-MoFox 官方文档的搜索与阅读功能，当用户询问框架功能、配置、插件开发、API 用法时使用"Restart Neo-MoFox for the changes to take effect.
{
"mcpServers": {
"mofox-docs": {
"command": "npx",
"args": ["-y", "mofox-docs-mcp"]
}
}
}Parameter | Description | Default |
| Documentation site URL |
|
| Request timeout |
|
| Cache duration for the index ( |
|
The same environment variables are also supported: MOFOX_DOCS_BASE_URL, MOFOX_DOCS_TIMEOUT, MOFOX_DOCS_CACHE_TTL.
npm run build # 编译到 dist/
npm start # 运行MIT
获取指定文档的完整正文（纯文本）。doc_id 形如 guides/deployment/deployment_guide，可从 search_docs / list_docs 结果的 id 字段获取。
Inputs: doc_id (required).
列出 Neo-MoFox 官方文档。按 section 排序，部署指南（guides/deployment）排最前，每篇含 title / description / preview / section。支持 offset 分页。
Inputs: limit, offset.
按关键词检索 Neo-MoFox 官方文档。查询按空白拆分为多个关键词，按「命中关键词数量」降序返回（命中全部关键词的文档排最前，至少命中一个即返回）。匹配标题 / 简介 / 正文预览 / 分类 / id / 路径。结果不含完整正文，需要全文时再用 get_doc 获取。
Inputs: query (required), limit, offset.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 3 tools
Each tool has a clearly distinct role: search_docs finds relevant docs by keywords, list_docs provides browsable pagination, and get_doc returns complete content for a specific doc. There is no meaningful overlap between them.
All tool names follow a clean verb_noun pattern: search_docs, list_docs, and get_doc. The slight singular/plural difference between docs and doc is natural and does not create confusion.
With only three tools, the server covers the core documentation workflow of search, browse, and retrieve without redundancy. This is an appropriately minimal surface for a docs-focused server.
The tools form a complete documentation lifecycle: discover content through search and listing, then fetch the full text via get_doc. There are no obvious gaps, as the server's documented purpose is fully supported.
Read-only MCP server over the developer docs for the portfolio's JSON API, MCP, A2A and NLWeb.
Read-only MCP server for the OrchestKit docs: full-text search + Markdown fetch. No auth.
Political Comms documentation MCP server: search docs, query the docs filesystem. No auth.
MCP server for querying Forkast documentation
