MCP.so
登录

Ollama Chat with MCP Server

@redbuilding

基本信息

分类

AI 与智能体

许可证

MIT

传输方式

stdio

作者

redbuilding

关于 Ollama Chat with MCP Server

This app demonstrates use of MCP server and client in a local model chat via Ollama that incorporates web search via Serper.

Ollama Chat with MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "ollama-chat-with-mcp": {
      "command": "python",
      "args": [
        "chat_frontend.py"
      ]
    }
  }
}

Ollama Chat with MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is 🔍 🤖 🌐 Ollama Chat with MCP?

🔍 🤖 🌐 Ollama Chat with MCP is a modern UI that integrates local (Ollama) and hosted LLMs with intelligent web search, content extraction, SQL, YouTube transcript analysis, HubSpot actions, Python data analysis, Canva design creation, Figma design inspection, Poe multi-model AI, and Codex code scaffolding — all via the Model Context Protocol (MCP). It is built for developers and power users who want to extend local and hosted models with external tools and data sources.

How to use 🔍 🤖 🌐 Ollama Chat with MCP?

Configure the backend with API keys for the desired providers (OpenAI, Anthropic, Google, etc.) and set up MCP services (web search, SQL, YouTube, HubSpot, Python, Canva, Figma, Poe, Codex). The frontend (React) provides a settings screen for API keys and model selection. Conversations and tasks persist in MongoDB.

Key features of 🔍 🤖 🌐 Ollama Chat with MCP

  • Smart web search with automatic content extraction
  • Multi‑provider LLM support (Ollama, OpenAI, Anthropic, Google, etc.)
  • Semantic memory and document knowledge base (ChromaDB)
  • Autonomous task execution with full MCP tool access
  • Canva, Figma, Poe, and Codex integrations
  • Conversation pinning, user profiles, and personalized AI

Use cases of 🔍 🤖 🌐 Ollama Chat with MCP

  • Design review and handoff using Figma integration
  • Automated data analysis and visualization from CSV uploads
  • Multi‑model content generation (text, image, video, audio)
  • Autonomous task workflows with planning, retries, and verification
  • Personalized AI assistance with user context and pinned conversations

常见问题

1

What LLMs does it support?

Local Ollama models plus hosted providers: OpenAI, Anthropic, Google, OpenRouter, Groq, and SambaNova — all selectable via a unified model picker.

2

What data storage is used?

MongoDB stores conversation history and user data; ChromaDB with nomic-embed-text embeddings powers semantic memory and the document knowledge base.

3

What MCP tools are available?

Fifty-five tools across nine modules: web search, SQL (read-only MySQL), YouTube transcripts, HubSpot, Python data analysis, Canva, Figma, Poe, and Codex workspace.

4

What are the runtime requirements?

Ollama for local models, API keys for hosted providers, a MongoDB instance, and optional tokens for Canva, Figma, and Poe. The backend (FastAPI) and frontend (React) must be running.

5

Is there a demo?

Yes, a demo video is available on X at https://x.com/redbuilding/status/2010124029936427450.

评论

AI 与智能体 分类下的更多 MCP 服务器