patent-mcp
AI-powered patent search and analysis across 220M+ global patents. Semantic search, prior art discovery, novelty/patentability reports, and patent content retrieval.
README
GoVeda Patent MCP Server
AI-powered patent intelligence for search & analysis via the Model Context Protocol (MCP).
Search and analyze 220M+ global patents with semantic search, prior art discovery, novelty/patentability reports, and full patent content retrieval — directly from Claude, ChatGPT, and other MCP-compatible AI assistants.
Endpoint: https://mcp.goveda.com/mcp
Transport: Streamable HTTP
Authentication: OAuth 2.0 (automatic browser sign-in)
Quick Start
Claude Desktop / claude.ai
Settings > Connectors > Add custom connector:
- Name: GoVeda Patent
- URL:
https://mcp.goveda.com/mcp
OAuth authentication opens automatically in your browser.
Claude Code
claude mcp add goveda-patent --transport http https://mcp.goveda.com/mcp
ChatGPT
Settings > Apps > Advanced > Enable Developer Mode. Add MCP server with URL https://mcp.goveda.com/mcp and OAuth authentication.
Cursor / Windsurf (mcp-remote bridge)
For clients that only support stdio transport:
{
"mcpServers": {
"goveda-patent": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.goveda.com/mcp"]
}
}
}
Requires Node.js installed. mcp-remote handles OAuth and protocol bridging.
VS Code / GitHub Copilot
Add to .vscode/mcp.json:
{
"servers": {
"goveda-patent": {
"type": "streamable-http",
"url": "https://mcp.goveda.com/mcp"
}
}
}
Tools
Search & Discovery
| Tool | Description | Cost |
|---|---|---|
search_patents |
Semantic search across 220M+ patents by technology description or concept | ~50-320 credits |
get_prior_art |
Find prior art for a known patent by publication number | ~690 credits |
get_search_status |
Poll async search/prior art progress and retrieve results | Free |
Reports
| Tool | Description | Cost |
|---|---|---|
generate_report |
AI-powered novelty/patentability assessment for an invention idea | 720 credits |
get_report_status |
Poll async report generation progress | Free |
get_report_summary |
Executive summary with patentability verdict | Free |
get_report_patent_analysis |
Detailed claim-by-claim comparison with cited prior art | Free |
Patent Content
| Tool | Description | Cost |
|---|---|---|
get_patent |
Retrieve specific patent sections by publication number | 1 credit |
batch_get_patents |
Retrieve multiple patents in one call | 1 credit/patent |
Available sections: abstract, claims, description, dates, parties, classifications, legal_status, citations, family, or all.
Account
| Tool | Description | Cost |
|---|---|---|
get_usage |
Check remaining credit balance and billing period | Free |
start_free_trial |
Activate 14-day trial with 10,000 credits | Free |
Example Prompts
Patent search:
"Search for patents about wireless charging for electric vehicles"
Prior art analysis:
"Find prior art for patent US-20230001234-A1"
Patent content:
"Show me the abstract and claims of US-11234567-B2"
Novelty report:
"Generate a novelty report for: A solar panel with self-cleaning nano-coating that activates upon detecting dust accumulation using piezoelectric sensors"
Credit check:
"Check my remaining credits"
Pricing
New users get a 14-day free trial with 10,000 credits — no credit card required. Credits are shared between the MCP server and the GoVeda web app.
See goveda.com/pricing for plans.
Important Notes
- Reports are AI-generated for informational purposes only. Professional legal review is recommended for patent decisions.
- All tools include MCP safety annotations (
readOnlyHint,destructiveHint,title). - Responses are truncated to 8,000 characters per field to limit data exposure.
Links
- GoVeda — Web application
- Documentation — Full MCP server documentation
- Privacy Policy
- Terms of Service
- Support
License
This repository contains documentation and configuration for the GoVeda Patent MCP Server. The server itself is a hosted service at mcp.goveda.com. See Terms of Service for usage terms.
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。