Smart Ai Bridge

Name: Smart Ai Bridge
Author: Platano78

@Platano78

关于 Smart Ai Bridge

Smart AI Bridge is a production-ready Model Context Protocol (MCP) server that orchestrates AI-powered development operations across multiple backends with automatic failover, smart routing, and advanced error prevention capabilities.

#smart-ai-bridge #ai-orchestration #development-tools

基本信息

分类

开发工具

传输方式

stdio

发布者

Platano78

提交者

Platano78

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "smart-ai-bridge": {
      "command": "node",
      "args": [
        "smart-ai-bridge.js"
      ],
      "cwd": ".",
      "env": {
        "LOCAL_MODEL_ENDPOINT": "http://localhost:1234/v1",
        "CLOUD_API_KEY_1": "your-cloud-api-key-1",
        "CLOUD_API_KEY_2": "your-cloud-api-key-2",
        "CLOUD_API_KEY_3": "your-cloud-api-key-3"
      }
    }
  }
}

工具

未检测到工具

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 Smart Ai Bridge?

Smart Ai Bridge is an enterprise-grade Model Context Protocol (MCP) server for Claude Desktop. It orchestrates AI‑powered development operations across multiple backends with automatic failover, smart routing, and advanced error prevention. It comes pre‑configured with a 4‑backend system (1 local model + 3 cloud AI backends) and is fully expandable to any AI provider.

How to use Smart Ai Bridge?

After installing dependencies (npm install) and running tests (npm test), add the server to Claude Code’s mcpServers configuration using a JSON block that specifies the node command, the script path, and environment variables for endpoint URLs and API keys. Then restart Claude Code to begin using the tools.

Key features of Smart Ai Bridge

Multi‑AI backend orchestration with intelligent routing and health‑aware failover.
Advanced fuzzy matching (exact → fuzzy → suggestions) reducing “text not found” errors by 80%.
19 total tools: 9 core plus 10 intelligent aliases for code review, file operations, and batch editing.
Enterprise security with 9.7/10 score, DoS protection, input validation, and audit trails.
Smart routing that auto‑selects backends based on task type (coding, analysis, large context).
Fully expandable architecture – add unlimited AI providers via the provided extension guide.

Use cases of Smart Ai Bridge

Route coding tasks (debugging, refactoring, game development) automatically to a coding‑specialist cloud backend.
Offload large file processing (>100 KB) to a local model with unlimited tokens.
Compare responses from different AI backends for the same prompt using the compare_endpoints tool.
Perform fuzzy‑matched file edits with configurable similarity thresholds and validation modes.
Analyze files across many paths with concurrent processing, security validation, and smart content routing.

FAQ from Smart Ai Bridge

What AI backends are supported?

The default setup uses LM Studio for the local backend and NVIDIA API for cloud backends, but you can configure any provider (OpenAI, Anthropic, Azure, AWS Bedrock, custom APIs, or local models via Ollama/vLLM). See the EXTENDING.md guide for details.

How does the smart routing work?

The system analyzes the prompt content and size: coding patterns (e.g., function, class, debug) route to the coding specialist; math/analysis patterns route to the analysis specialist; prompts >50,000 characters or files >100 KB route to the local backend; everything else defaults to the highest‑priority cloud backend.

What is the fuzzy matching threshold and can I adjust it?

The fuzzy matching engine uses a configurable similarity threshold from 0.1 to 1.0 (default 0.8). You can also set validation mode to strict, lenient, or dry_run per operation.

Is Smart Ai Bridge production ready?

Yes. It has 100% test coverage, enterprise‑grade reliability, and is MIT licensed. The security score is 9.7/10 with DoS protection, input validation, and complete audit logging.

How do I add more backends beyond the default four?

The system is fully expandable. Follow the step‑by‑step instructions in EXTENDING.md to integrate OpenAI, Anthropic, Azure, AWS Bedrock, or any custom API or local model endpoint.