NexusAPI
NexusAPI provides a suite of 18 compute tools for AI agents, including web scraping, sandboxed Python code execution, and NLP processing. It enables agents to perform complex tasks like sentiment analysis, image manipulation, and data extraction through a unified interface.
README
NexusAPI — 18 Compute Tools for AI Agents
Web scraping, sandboxed code execution, ML inference, and 15 more tools. 250 free credits on signup.
Free Tools (No API Key Required)
Try these right now — no signup, no API key:
# Convert any URL to clean markdown
curl "https://marble-appendix-accurately-mark.trycloudflare.com/v1/try/markdown?url=https://news.ycombinator.com"
# Scrape any URL to text
curl "https://marble-appendix-accurately-mark.trycloudflare.com/v1/try/scrape?url=https://example.com"
# Analyze sentiment
curl "https://marble-appendix-accurately-mark.trycloudflare.com/v1/try/sentiment?text=AI+agents+are+amazing"
Install the SDK
pip install git+https://github.com/ruizmr/nexusapi-mcp.git
Quick Start (3 lines)
from nexusapi import NexusAPI
api = NexusAPI() # auto-registers, instant API key, 250 free credits
page = api.scrape("https://news.ycombinator.com")
result = api.execute("import math; print(math.factorial(20))")
sentiment = api.sentiment("Agent-to-agent commerce is the future")
Agent Self-Registration
curl -X POST https://marble-appendix-accurately-mark.trycloudflare.com/v1/agent/register \
-H "Content-Type: application/json" \
-d '{"agent_name": "my-agent", "contact_email": "agent@example.com"}'
# Returns: {"api_key": "nx_...", "credits": 250}
All 18 Tools
| Category | Endpoint | Credits | Description |
|---|---|---|---|
| Web | /v1/web/scrape |
5 | URL → clean text/markdown/links |
| Web | /v1/web/scrape/batch |
25 | Batch scrape up to 5 URLs |
| Code | /v1/exec/python |
20 | Sandboxed Python execution |
| NLP | /v1/text/sentiment |
10 | Sentiment analysis |
| NLP | /v1/text/summarize |
10 | Text summarization |
| NLP | /v1/text/keywords |
5 | Keyword extraction |
| NLP | /v1/text/entities |
10 | Named entity recognition |
| NLP | /v1/text/readability |
1 | Readability scoring |
| ML | /v1/ml/zero-shot |
10 | Zero-shot classification |
| Image | /v1/image/resize |
5 | Image resize/optimize |
| Image | /v1/image/blur |
3 | Image blur/effects |
| Data | /v1/data/hash |
1 | SHA256/MD5 hashing |
| Data | /v1/data/encode |
1 | Base64/URL encoding |
| Data | /v1/data/regex |
1 | Regex matching |
| Util | /v1/util/qr |
2 | QR code generation |
| Util | /v1/util/uuid |
1 | UUID generation |
| Util | /v1/util/timestamp |
1 | Timestamp conversion |
MCP Integration
{
"mcpServers": {
"nexusapi": {
"url": "https://marble-appendix-accurately-mark.trycloudflare.com/.well-known/mcp.json"
}
}
}
Framework Integration
CrewAI:
from crewai_tools import tool
from nexusapi import NexusAPI
api = NexusAPI()
@tool
def scrape(url: str) -> str:
return api.scrape(url)["content"]
LangChain / smolagents / Agno: Same pattern — wrap any SDK method in a tool decorator.
Payment
- Free: 250 credits on signup (enough for 50 scrapes or 12 code executions)
- Top up: USDC on Base L2 → auto-credited in 30 seconds
- Pricing: 1 credit ≈ $0.001
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。