patent-search
About
This Claude skill enables patent searching across 76M+ patents via BigQuery for fast, zero-setup global searches, or through the PatentsView API for detailed US patent metadata. Developers can quickly test it using the provided Python script with pre-configured credentials. Choose BigQuery for worldwide coverage or PatentsView for rich US patent details like inventors and citations.
Documentation
Patent Search Skill
Two powerful patent search methods:
- BigQuery Search (Recommended) - 76M+ worldwide patents, zero local storage
- PatentsView API - Detailed US patent metadata (inventors, assignees, classifications, citations)
FOR CLAUDE: All files and dependencies installed.
- Go directly to Quick Test section
- Script at:
.claude/skills/patent-search/bigquery_search.py - Run from skill directory
- Windows: Use cmd syntax (dir, set, &&)
Quick Test
# Windows
cd ".claude\skills\patent-search"
set GOOGLE_APPLICATION_CREDENTIALS=%APPDATA%\\gcloud\\application_default_credentials.json
python bigquery_search.py search "voice biometric" 5
# Linux/macOS
cd .claude/skills/patent-search
export GOOGLE_APPLICATION_CREDENTIALS="$HOME/.config/gcloud/application_default_credentials.json"
python bigquery_search.py search "voice biometric" 5
Expected: 5 patent results in ~4 seconds
Quick Start
BigQuery Search
# Keyword search (2-3 keywords for best results)
python bigquery_search.py search "voice biometric authentication" 20
# Get specific patent (hyphenated format: US-XXXXXXX-XX)
python bigquery_search.py get US-12424224-B2
# CPC classification search
python bigquery_search.py cpc G10L 15
PatentsView API
Choosing the Right Method
Use BigQuery When:
- Quick keyword search needed
- Worldwide patents (not just US)
- Fast results (3-4 seconds)
- Zero local storage
- CPC classification search
- Budget-conscious (free tier: 1TB/month)
Use PatentsView When:
- Need detailed US patent metadata
- Searching by inventor/assignee
- Citation analysis required
- Complex boolean queries
- Exact field matching
- Patent family analysis
Combined Workflow
- Start with BigQuery (broad keyword search)
- Identify relevant patents and CPC codes
- Switch to PatentsView (detailed metadata/citations)
- Export final results
Example:
# Step 1: BigQuery broad search
python bigquery_search.py search "voice biometric authentication" 20
# Step 2: Found CPC G10L17, search more
python bigquery_search.py cpc G10L17 50
# Step 3: Use PatentsView for inventor/assignee analysis
Instructions for Claude
When user requests patent searches:
- Understand Goal: Technology, time period, prior art vs competitive analysis, US vs worldwide
- Check Dependencies: Verify BigQuery/PatentsView setup
- Choose Method: Default BigQuery for broad, PatentsView for detailed
- Optimize Queries:
- BigQuery: 2-3 keywords, simplify if zero results, use CPC codes
- PatentsView: Verify API key, construct JSON queries, handle pagination
- Present Results: Parse JSON, highlight key info, provide Google Patents URLs
- Offer Next Steps: Suggest refinements, related classifications, citation analysis
Common Use Cases
Prior Art Search
- BigQuery keyword search
- Identify CPC codes
- BigQuery CPC search
- PatentsView citation analysis
- Document findings
Competitive Intelligence
- PatentsView search by assignee
- Filter by date range
- Group by CPC
- Identify key inventors
- Trend report
Technology Landscape
- BigQuery CPC search worldwide
- Analyze by country/date
- Identify patent families
- PatentsView US details
- Summary report
Freedom to Operate
- BigQuery keyword + CPC search
- Filter by jurisdiction/active status
- PatentsView claim analysis
- Review forward citations
- Risk assessment
Performance & Coverage
| Method | Patents | Coverage | Speed | Cost | Storage |
|---|---|---|---|---|---|
| BigQuery | 76M+ | Worldwide | 3-4s | Free* | 0GB |
| PatentsView | 9.2M | US only | 1-3s | Free | 0GB |
*Free tier: 1TB queries/month
Quick Reference
BigQuery Commands
python bigquery_search.py search "query" <limit>
python bigquery_search.py get <PATENT-NUMBER>
python bigquery_search.py cpc <CODE> <limit>
Common CPC Codes
| Code | Technology |
|---|---|
| G10L | Speech analysis/synthesis |
| G10L15 | Speech recognition |
| G10L17 | Speaker recognition/verification |
| G06F21 | Security arrangements |
| G06N | Computing models |
Troubleshooting
FOR CLAUDE: Only run diagnostics if Quick Test fails.
| Problem | Solution |
|---|---|
| BigQuery auth fails | gcloud auth application-default login |
| No module google.cloud | pip install google-cloud-bigquery db-dtypes |
| Zero results | Simplify query (2-3 keywords max) |
| Patent get fails | Use hyphenated format: US-XXXXX-XX |
| PatentsView 403 | Set API key environment variable |
| Rate limit (429) | Wait 60 seconds (PatentsView: 45 req/min) |
Quick Install
/plugin add https://github.com/RobThePCGuy/Claude-Patent-Creator/tree/main/patent-searchCopy and paste this command in Claude Code to install this skill
GitHub 仓库
Related Skills
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
csv-data-summarizer
MetaThis skill automatically analyzes CSV files to generate comprehensive statistical summaries and visualizations using Python's pandas and matplotlib/seaborn. It should be triggered whenever a user uploads or references CSV data without prompting for analysis preferences. The tool provides immediate insights into data structure, quality, and patterns through automated analysis and visualization.
llamaindex
MetaLlamaIndex is a data framework for building RAG-powered LLM applications, specializing in document ingestion, indexing, and querying. It provides key features like vector indices, query engines, and agents, and supports over 300 data connectors. Use it for document Q&A, chatbots, and knowledge retrieval when building data-centric applications.
