NexusBridge MCP Server
Provides AI agents access to over 200 AI models and 10 service categories via a unified, metered API with transparent per-call pricing.
README
NexusBridge MCP Server
A Model Context Protocol server that gives AI agents access to 200+ AI models and 10 service categories through a unified, metered API. One API key, one integration, transparent per-call pricing.
API Base: https://syntss.com
Smithery: smithery.ai/servers/guhank/nexusbridge-mcp-server
GitHub: github.com/guhank/nexusbridge-mcp-server
Tools
| Tool | Description |
|---|---|
nexusbridge_catalog |
Retrieve the full service catalog with slugs, descriptions, per-call prices, and your current credit balance. Call this first to discover available services. |
nexusbridge_execute |
Execute any NexusBridge service by its slug. Pass service-specific parameters. Credits are deducted automatically. Returns the upstream provider's response. |
nexusbridge_balance |
Check your current credit balance (USD) and the number of active services available. Use before expensive operations to verify sufficient funds. |
nexusbridge_catalog
Returns every available service with its slug, name, category, description, and per-call price in USD. Also returns your current credit balance.
Parameters: None
Example response:
{
"services": [
{ "slug": "llm-chat", "name": "LLM Chat", "category": "ai", "pricePerCall": 0.005 },
{ "slug": "web-search", "name": "Web Search", "category": "search", "pricePerCall": 0.01 }
],
"creditBalance": 24.50
}
nexusbridge_execute
Execute a service by passing its slug and service-specific parameters.
Parameters:
| Parameter | Type | Required | Description |
|---|---|---|---|
slug |
string | Yes | Service slug from the catalog (e.g., llm-chat, web-search, sentiment) |
params |
object | No | Service-specific parameters (see examples below) |
Examples:
// LLM Chat — use any of 200+ models
{ "slug": "llm-chat", "params": { "messages": [{"role": "user", "content": "Explain quantum computing"}], "model": "deepseek/deepseek-chat-v3-0324" } }
// Sentiment Analysis
{ "slug": "sentiment", "params": { "text": "I love this product! Best purchase I've ever made." } }
// Web Search
{ "slug": "web-search", "params": { "query": "latest AI news March 2026" } }
// Text Summarization
{ "slug": "summarize", "params": { "text": "Long article text here..." } }
// Translation
{ "slug": "translation", "params": { "text": "Hello world", "to": "es" } }
// Code Generation
{ "slug": "llm-code", "params": { "prompt": "Write a Python function to calculate fibonacci numbers" } }
// Image Generation
{ "slug": "image-gen", "params": { "prompt": "A serene mountain landscape at sunset" } }
// Text Embedding
{ "slug": "text-embedding", "params": { "text": "semantic search query" } }
Example response:
{
"success": true,
"service": "llm-chat",
"charged": 0.005,
"remainingBalance": 24.495,
"result": {
"choices": [{ "message": { "content": "Quantum computing uses..." } }]
}
}
nexusbridge_balance
Returns your credit balance and the count of active services.
Parameters: None
Example response:
{
"creditBalance": 24.50,
"serviceCount": 10
}
Available Services
| Service | Slug | Price/Call | Description |
|---|---|---|---|
| LLM Chat | llm-chat |
$0.005 | Chat completions with GPT-4o, Claude, Gemini, DeepSeek, Llama, and 200+ models via OpenRouter |
| Code Generation | llm-code |
$0.008 | Code generation, refactoring, and completion optimized for development tasks |
| Image Generation | image-gen |
$0.05 | Create images from text prompts using DALL-E, SDXL, and Flux models |
| Text Embedding | text-embedding |
$0.001 | Vector embeddings for RAG pipelines, semantic search, and clustering |
| Web Search | web-search |
$0.01 | Structured web search results with titles, URLs, and snippets |
| Document Parser | doc-parser |
$0.02 | Extract text and structured data from PDFs, images, and documents |
| Sentiment Analysis | sentiment |
$0.003 | Classify text sentiment (positive/negative/neutral) with confidence scores and emotion detection |
| Translation | translation |
$0.005 | Neural machine translation across 100+ languages |
| Summarization | summarize |
$0.008 | Configurable-length text summaries from long documents |
| Code Execution | code-exec |
$0.015 | Run code in sandboxed Python, Node.js, and Go environments |
Installation
Smithery (Recommended)
Install directly from Smithery with one click. Configure your NexusBridge API key when prompted.
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"nexusbridge": {
"command": "npx",
"args": ["-y", "nexusbridge-mcp-server", "--stdio"],
"env": {
"NEXUSBRIDGE_API_KEY": "nb_sk_your_key_here"
}
}
}
}
Cursor / Windsurf / Other MCP Clients
Use the same configuration pattern — set NEXUSBRIDGE_API_KEY as an environment variable and run via npx -y nexusbridge-mcp-server --stdio.
Docker
docker build -t nexusbridge-mcp .
docker run -p 8000:8000 \
-e NEXUSBRIDGE_API_KEY=nb_sk_your_key \
nexusbridge-mcp
Manual
git clone https://github.com/guhank/nexusbridge-mcp-server.git
cd nexusbridge-mcp-server
npm install
NEXUSBRIDGE_API_KEY=nb_sk_your_key npm start
Usage
-
Get an API key — Purchase credits at syntss.com. Credits start at $10. You'll receive an
nb_sk_...API key. -
Install the MCP server — Use Smithery, Claude Desktop config, or run manually (see Installation above).
-
Discover services — Your AI agent calls
nexusbridge_catalogto see all available services, pricing, and your credit balance. -
Execute services — Call
nexusbridge_executewith the service slug and parameters. Credits are deducted per call. -
Monitor spending — Call
nexusbridge_balanceto check remaining credits at any time.
Typical Agent Workflow
Agent → nexusbridge_catalog → "I see 10 services, balance is $24.50"
Agent → nexusbridge_execute(llm-chat) → "Got response from DeepSeek, charged $0.005"
Agent → nexusbridge_execute(sentiment)→ "Text is positive (0.94 confidence), charged $0.003"
Agent → nexusbridge_balance → "Balance is $24.492"
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
NEXUSBRIDGE_API_KEY |
Yes | — | Your NexusBridge API key (starts with nb_sk_) |
NEXUSBRIDGE_BASE_URL |
No | https://syntss.com |
Base URL of the NexusBridge API |
PORT |
No | 8000 |
HTTP port for Smithery/hosted mode |
Transport Modes
| Mode | Flag | Use Case |
|---|---|---|
| HTTP | (default) | Smithery hosted deployment, remote MCP clients |
| stdio | --stdio |
Claude Desktop, Cursor, local MCP clients |
Getting API Credits
| Tier | Credits | Approx. LLM Calls |
|---|---|---|
| Starter | $10 | ~2,000 |
| Builder | $25 | ~5,000 |
| Pro | $50 | ~10,000 |
| Scale | $100 | ~20,000 |
Purchase at syntss.com or via Stripe. Credits never expire.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。