convez.ai
MCP ServerIntegração & APIsAvançado
Palm1r

LLMQore

Palm1r/llmqore

Biblioteca Qt/C++ para integração LLM e servidores MCP com transporte unificado

Ver no GitHub
BomPluginScore · de 0 a 100

Biblioteca Qt/C++ para integração com provedores de LLM na nuvem e locais, criação de servidores e clientes MCP com suporte a streaming unificado. Inclui MCP Bridge, ferramenta CLI que agrega múltiplos servidores MCP e re-expõe por HTTP ou stdio.

Instalação

$npx convez-ai Palm1r/llmqore

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
53
Forks
16
Acompanhando
2
Saúde
42%
Último commit
há 1 mês
Versão
v0.8.0
Issues abertas
0
Contribuidores
2

Casos de uso

  • Criar aplicações desktop C++ que integram múltiplos provedores de LLM
  • Construir servidores MCP que expõem tools e recursos para Claude Desktop
  • Agregar e adaptar transporte de múltiplos servidores MCP (stdio para HTTP ou vice-versa)

Veredito

Solução robusta para desenvolvedores C++ que precisam integrar LLMs e MCP em aplicações desktop. Cobertura ampla de provedores e transporte flexível tornam-na versátil, mas requer expertise em C++ e Qt para explorar todo o potencial.

README

A tradução para o português deste README ainda está na fila.

LLMQore

Build and Test GitHub Tag

Qt/C++ library for working with cloud and local LLM providers, create MCP servers and clients, download and using library MCP Bridge.

LLM clients — unified streaming API across all providers:

auto *client = new LLMQore::ClaudeClient(url, apiKey, model, this);
client->ask("What is Qt?", cb);

MCP server — expose tools, resources and prompts over stdio or HTTP:

// stdio (stdin/stdout, e.g. for Claude Desktop)
auto *transport = new LLMQore::McpStdioServerTransport(&app);

// or Streamable HTTP
auto *transport = new LLMQore::McpHttpServerTransport({.port = 8080, .path = "/mcp"}, &app);

auto *server = new LLMQore::McpServer(transport, cfg, &app);
server->addTool(new MyTool(server));
server->start();

MCP client — connect to MCP servers and bind their tools into LLM clients:

// Add servers one by one
client->tools()->addMcpServer({.name = "filesystem", .command = "npx",
    .arguments = {"-y", "@modelcontextprotocol/server-filesystem", "/home/user"}});

// Or load from a JSON config
client->tools()->loadMcpServers(QJsonDocument::fromJson(configData).object());

loadMcpServers accepts:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/user"]
    }
  }
}

See Quick Start for complete examples.

MCP Bridge

A standalone CLI tool built on llmqore that aggregates multiple MCP servers (stdio or SSE) and re-exposes them either behind a single HTTP/SSE endpoint or as one stdio server — useful when the upstreams and the client disagree on transport.

mcp-bridge bridge.json              # HTTP endpoint
mcp-bridge --stdio bridge.json      # stdio (for Claude Desktop and friends)

Config uses the familiar mcpServers schema:

{
  "port": 8808,
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/user"]
    }
  }
}

Prebuilt binaries for Linux/macOS/Windows (with Qt runtime bundled) are published to GitHub Releases. See MCP Bridge docs for full usage, config reference, and build instructions.

Supported Providers

ProviderClient classStreamingToolsThinking
Anthropic ClaudeClaudeClient✓✓✓
OpenAI (Chat Completions)OpenAIClient✓✓✓
OpenAI (Responses API)OpenAIResponsesClient✓✓✓
OllamaOllamaClient✓✓✓
Google AIGoogleAIClient✓✓✓
MistralMistralClient✓✓✓
llama.cppLlamaCppClient✓✓✓
DeepSeekOpenAIClient✓✓✓

MCP (Model Context Protocol)

Client and server implementation of the MCP 2025-11-25 spec:

  • Transports: stdio, Streamable HTTP
  • Server: tools, resources, resource templates, prompts, completions, sampling, elicitation
  • Client: tools, resources, prompts, completions, sampling, elicitation, roots

See MCP Protocol Coverage for the full spec-conformance matrix.

Requirements

  • C++20
  • Qt 6.5+
  • CMake 3.21+

Documentation

Support

  • Report Issues: open an issue on GitHub
  • Contribute: pull requests with bug fixes or new features are welcome
  • Spread the Word: star the repository and share with fellow developers
  • Financial Support:
    • Paypal: paypal page
    • Bitcoin (BTC): bc1qndq7f0mpnlya48vk7kugvyqj5w89xrg4wzg68t
    • Ethereum (ETH): 0xA5e8c37c94b24e25F9f1f292a01AF55F03099D8D
    • Litecoin (LTC): ltc1qlrxnk30s2pcjchzx4qrxvdjt5gzuervy5mv0vy
    • USDT (TRC20): THdZrE7d6epW6ry98GA3MLXRjha1DjKtUx

License

MIT — see LICENSE.

Ver o original no GitHub

Bom para

Relacionados