Yantrix MCP
Dynamic MCP server — 30+ tools across fact verification, agent memory, Indian NLP, contract risk, security threat modelling, sales call intelligence and more. x402/USDC micropayments on Base.
README
@yantrixai/mcp
Dynamic MCP server for the Yantrix API portfolio. One install gives any MCP-compatible AI agent access to 30+ tools across research, memory, language, legal, security, intelligence, and more.
What makes it dynamic
Unlike static MCP packages, @yantrixai/mcp fetches its tool list from registry.yantrix.ai on startup. When new APIs are added to the Yantrix portfolio, they appear automatically on your next MCP host restart — no package update needed.
Install
npm install -g @yantrixai/mcp
Or use directly with npx:
npx @yantrixai/mcp
Configuration
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"yantrix": {
"command": "npx",
"args": ["@yantrixai/mcp"],
"env": {
"X_PAYMENT_HEADER": "<your-x402-payment-proof>",
"DEV_MODE": "false"
}
}
}
}
Development / Testing (no payment required)
{
"mcpServers": {
"yantrix": {
"command": "npx",
"args": ["@yantrixai/mcp"],
"env": {
"DEV_MODE": "true"
}
}
}
}
Environment Variables
| Variable | Description | Default |
|---|---|---|
X_PAYMENT_HEADER |
x402 payment proof for USDC micropayments | — |
DEV_MODE |
Skip payment validation (local testing only) | false |
YANTRIX_REGISTRY_URL |
Custom registry URL | https://registry.yantrix.ai/mcp-registry.json |
Available Tools
Tools are loaded dynamically from the registry. Current categories:
| Category | Tools |
|---|---|
research |
verify_claim, deep_verify_claim, batch_verify |
memory |
write_memory, read_memory, search_memory |
language |
transliterate, indic_sentiment, detect_code_switching, indic_ner |
legal |
scan_contract, scan_clause |
security |
model_threats, check_compliance |
intelligence |
score_trend, compare_trends, scan_competitor |
education |
generate_exam |
content |
rewrite_in_style |
india |
company_lookup, gst_intelligence, lookup_ifsc |
sales |
analyze_call, extract_crm_data |
utility |
validate_phone, detect_timezone, convert_timezone, find_timezone_overlap |
infrastructure |
check_quota, consume_quota, track_api_event |
reasoning |
argue_claim, detect_fallacies |
Payment
All tools use x402 micropayments in USDC on Base Mainnet.
Costs range from $0.0001 (event tracking) to $0.025 (sales call analysis).
To get a payment header, visit x402.org or use the x402 client SDK.
How it works
1. npx @yantrixai/mcp starts
2. Fetches https://registry.yantrix.ai/mcp-registry.json
3. Registers all tools dynamically
4. MCP host calls any tool
5. @yantrixai/mcp routes call to correct Yantrix API
6. Response returned to agent
The registry is cached for 5 minutes. New tools appear on next restart.
Adding new APIs (for maintainers)
No npm republish needed. Just update main.py in the mcp-registry service:
# Add to TOOLS list in mcp-registry/main.py
{
"name": "new_tool_name",
"service": "newapi",
"endpoint": "https://newapi.yantrix.ai/endpoint",
"method": "POST",
"description": "What this tool does.",
"input_schema": {...},
"cost_usdc": "0.005",
"category": "category",
}
Deploy the registry → all users get the new tool on next restart.
Links
- Registry: https://registry.yantrix.ai
- Docs: https://yantrix.ai
- x402: https://x402.org
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。