garden.stanislav.svelte-llm由 svelte-llm.stanislav.garden 發布官方發布者v1.0.0更新於 Oct 24, 2026
An MCP server that provides access to Svelte 5 and SvelteKit documentation
遠端 MCP 端點，讓助理依需求瀏覽並讀取 Svelte 5 與 SvelteKit 文件章節。
Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。
把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。
閱讀 https://sourceweft.com/mcp/garden.stanislav.svelte-llm%2Fsvelte-llm-mcp/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。把它新增到你客戶端的 mcpServers 設定中。
{
"mcpServers": {
"svelte-llm": {
"type": "http",
"url": "https://svelte-llm.stanislav.garden/mcp/mcp"
}
}
}LLM presets and MCP for Svelte 5 and SvelteKit. Visit the site at svelte-llm.stanislav.garden.
This service provides an MCP (Model Context Protocol) endpoint for use with AI assistants:
https://svelte-llm.stanislav.garden/mcp/mcphttps://svelte-llm.stanislav.garden/mcp/sse.env file with the content:
public_repo permissions.docker-compose upYou can also visit http://localhost:5173/admin to see all the "hidden" endpoints (default password = "secret")
You can also use GitHub Copilot in Agent mode to try the agent locally, see .vscode/mcp.json for info on how that works.
OG image from https://dynamic-og-image-generator.vercel.app/
來源：README.md，提交 0835468
Lists all available Svelte 5 and SvelteKit documentation sections in a structured format. Returns sections as a list of "* title: [section_title], path: [file_path]" - you can use either the title or path when querying a specific section via the get_documentation tool. Always run list_sections first for any query related to Svelte development to discover available content.
無需輸入參數。
Retrieves full documentation content for Svelte 5 or SvelteKit sections. Supports flexible search by title (e.g., "$state", "routing") or file path (e.g., "docs/svelte/state.md"). Can accept a single section name or an array of sections. Before running this, make sure to analyze the users query, as well as the output from list_sections (which should be called first). Then ask for ALL relevant sections the user might require. For example, if the user asks to build anything interactive, you will need to fetch all relevant runes, and so on.
section必填已滿足適用檢查項的 100%
該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。
GITHUB_TOKEN=DB_URL=postgres://admin:admin@localhost:5432/dbnvm usenpm inpm run devNODE_TLS_REJECT_UNAUTHORIZED=0 NODE_OPTIONS="--insecure-http-parser" npx @modelcontextprotocol/inspector