MCP Servers

vision-mcp-worker

MCP server deployed on Cloudflare Worker that provides AI with image understanding via the describe_image tool, plus an Apple-style model status dashboard.

README

vision-mcp-worker

部署在 Cloudflare Worker 上的 MCP 服务器，给 AI 提供"看图"能力（describe_image 工具），附带一个果味模型状态展示页。

MCP 端点：https://<your-domain>/mcp（Streamable HTTP，建议用 Cloudflare Access Service Auth 保护）
状态页：https://<your-domain>/（模型与健康状态一览）
详细接入文档：见 MCP_SETUP.md
部署配置模板：见 wrangler.toml.example（真实 wrangler.toml 不提交）

模型配置与切换（环境变量）

模型完全通过环境变量配置，不用改代码。 修改后 pnpm run deploy 生效。

1. Workers AI 模型（默认，免费额度）

在 wrangler.toml 的 [vars] 里设置 VISION_MODEL：

[vars]
VISION_MODEL = "@cf/meta/llama-3.2-11b-vision-instruct"

默认回退模型为 Meta Llama 3.2 11B Vision（稳定、抠文字/数字细节强、便宜）。

2. 第三方 OpenAI 兼容模型（可选，优先于 Workers AI）

任意 OpenAI 兼容的视觉模型提供商（OpenAI / DeepSeek / 智谱 / 中转站 / Ollama 等）：

[vars]
OPENAI_API_BASE = "https://api.example.com/v1"   # 提供商 OpenAI 兼容端点
OPENAI_API_MODEL = "gpt-4o-mini"                  # 第三方模型名（不填则复用 VISION_MODEL）

# API Key 用 secret 存，不要写进仓库：
npx wrangler secret put OPENAI_API_KEY

优先级：配置了 OPENAI_API_BASE + OPENAI_API_KEY → 走第三方（chat/completions，图片以 image_url 传入）；否则走 Workers AI 的 VISION_MODEL。

3. 切换示例

想要的效果	配置
用 Workers AI 的 llama（回退默认）	`VISION_MODEL` 设为 llama，注释掉 `OPENAI_API_BASE`
用第三方模型	设置 `OPENAI_API_BASE` + `OPENAI_API_MODEL` + `OPENAI_API_KEY`
换其他第三方模型	改 `OPENAI_API_MODEL` 即可，不用动代码

注意事项

OPENAI_API_KEY 是 secret，存于 Cloudflare（wrangler secret），状态页只显示"已配置/未配置"
第三方中转若不稳定，健康检查会如实显示 DEGRADED，属正常现象
多模态模型才能做图片识别；纯文本模型（如 kimi-k2 系列）不能用于 describe_image

开发

pnpm install     # 依赖
pnpm run deploy  # 部署到 Cloudflare
pnpm test        # 测试

文件	说明
`src/index.js`	Worker 入口：MCP 端点 / 状态页 / 健康检查 / 模型路由
`wrangler.toml.example`	部署配置模板（真实配置见本地 `wrangler.toml`）
`tools/vision.js`	本地截图识别脚本
`tools/dashboard-demo.html`	测试页（仪表盘）
`MCP_SETUP.md`	完整接入与排障文档

Recommended Servers

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official

Featured

TypeScript

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official

Featured

Local

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official

Featured

TypeScript

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official

Featured

Python

E2B

Using MCP to run code via e2b.

Official

Featured

Neon Database

MCP server for interacting with Neon Management API and databases

Official

Featured

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official

Featured