ホーム/リーガル&契約/Claude-Patent-Creator
C

Claude-Patent-Creator

by @RobThePCGuyv
4.3(20)

Claude Patent Creator は、Claude Code 専用に設計されたAIベースの特許作成・分析システムです。RAG技術を統合しており、MPEP、USCなどの特許法規を迅速に検索できるほか、BigQueryを活用して数千万件の特許にアクセスし、先行技術調査を行います。また、システムは自動コンプライアンスチェック(例:35 USC 112)を提供し、グラフ生成もサポートしており、特許専門家の作業効率を大幅に向上させることを目指しています。

patentailegaltechragcomplianceGitHub
インストール方法
git clone https://github.com/RobThePCGuy/Claude-Patent-Creator.git
compare_arrows

Before / After 効果比較

1
使用前

従来の特許出願プロセスは時間がかかり、大量の規制文書の手動確認、先行技術調査に数週間を要し、請求項の適合性に関する手動審査は間違いを起こしやすく、効率が低い。

使用後

AIシステムを活用することで、規制検索は数秒で応答し、先行技術検索は数億件の特許をカバーし、自動適合性チェックは具体的なフィードバックを提供し、出願期間を大幅に短縮し、特許明細書作成の品質を向上させる。

SKILL.md

Claude Patent Creator

MIT License Python 3.9+ MCP Server PyTorch Status

An AI-powered patent creation and analysis system for Claude Code.

I built this because I needed to file a patent myself. I used AI to build the system, used the system to file the patent, and it worked. Now it's open source so anyone can use it, whether you're a developer exploring AI tooling or a patent professional looking to speed up your workflow.

In plain terms, this tool lets you:

  • Search patent regulations instantly. Ask a question about MPEP, 35 USC, 37 CFR, EPC, or PCT rules and get the relevant sections in under a second, with citations.
  • Find prior art across 100M+ patents. Search Google's BigQuery patent database by keywords, CPC/IPC codes, or full-text queries. Link related patents across jurisdictions with family search.
  • Check your claims for compliance. Run your draft claims through automated analysis for USPTO (35 USC 112(b)) or EPO (Art. 84 EPC) and get specific feedback on definiteness, two-part form, and structure.
  • Review your full application. Specification adequacy, formalities, required sections — checked against USPTO, EPO, or PCT standards.
  • Search EP patents with full text. Get full claims and description text for European patents via the EPO OPS API (not available in BigQuery).
  • Generate patent diagrams. Block diagrams, flowcharts, and system architectures in patent style, no design tools needed.
  • Draft a complete patent application. A guided workflow that walks you through the whole process, from invention disclosure to filing-ready documents — for USPTO or EPO filing.

Table of Contents


Quick Start

Pick the path that fits your setup. All three get you to the same place.

Option A: Claude Code Plugin (Easiest)

If you're already using Claude Code, this is the fastest way in:

# Add the marketplace and install
/plugin marketplace add RobThePCGuy/Claude-Patent-Creator
/plugin install claude-patent-creator-standalone@claude-patent-creator

# Run setup
/claude-patent-creator-standalone:setup-patent-system

Option B: One-Line Install

pip install git+https://github.com/RobThePCGuy/Claude-Patent-Creator.git && patent-creator setup

This handles everything automatically: installs dependencies, detects your GPU, downloads MPEP PDFs (~500MB), builds the search index, and registers the MCP server with Claude Code. Restart Claude Code when it finishes.

Option C: Manual Install

git clone https://github.com/RobThePCGuy/Claude-Patent-Creator.git
cd Claude-Patent-Creator

# Optional: use a virtual environment
python -m venv venv
source venv/bin/activate  # Linux/macOS
venv\Scripts\activate     # Windows

pip install -e .
patent-creator setup

Verify It Worked

After any install path, run:

patent-creator health

You should see a status report showing which components are ready. If something's off, the output will tell you what to fix.


What Can I Actually Do With This?

Here are some real examples. You can type these directly in Claude Code and the right skill or tool kicks in automatically.

What you want to doWhat to sayWhat happens
Find the MPEP rule on claim definiteness"Search MPEP for claim definiteness requirements"Hybrid search returns the most relevant MPEP sections with citations
Look for prior art"Search for patents about neural network training filed in 2024"BigQuery searches 100M+ patents and returns matching results
Check your claims"Review these claims for 35 USC 112(b) compliance"Automated analysis flags indefinite terms, missing antecedent basis, structural issues
Review a full application/full-reviewRuns claims + specification + formalities checks in parallel
Create a patent from scratch/create-patentGuided 6-phase workflow, takes 55-80 min, produces a complete filing package
Generate a diagram"Create a block diagram showing the system architecture"Generates a patent-style Graphviz diagram
Search prior art thoroughly"Conduct a prior art search for [your invention]"Automated novelty and freedom-to-operate analysis

How It Works

The system has two modes that can work independently or together:

MCP Server is the engine. It exposes 20+ tools that any MCP-compatible client (Claude Code, Claude Desktop, etc.) can call programmatically. These tools handle search, analysis, and diagram generation.

Claude Code Plugin adds the interactive layer. Skills activate automatically based on what you're doing. Agents handle long-running tasks in the background. Slash commands give you quick access to common workflows.

Under the hood, patent regulation search uses a hybrid approach: FAISS vector search finds semantically similar content, BM25 lexical search catches exact terminology matches, and a cross-encoder reranker sorts the combined results by relevance. Patent search goes through Google BigQuery's public patent dataset.

You (Claude Code) ──> MCP Server ──> Search / Analysis / Diagrams
                           │
            ┌──────────────┼──────────────┐
            v              v              v
     MPEP/USC/CFR     BigQuery        Graphviz
     (hybrid RAG)    (100M+ patents)   (diagrams)

Installation Options

  1. Installs Python package dependencies
  2. Detects your hardware (NVIDIA GPU, Apple Silicon, or CPU-only)
  3. If a GPU is detected, uninstalls CPU-only PyTorch and installs the GPU version
  4. Restarts the setup process with GPU-enabled PyTorch
  5. Downloads MPEP, 35 USC, and 37 CFR PDFs (~500MB) from the USPTO
  6. Builds the hybrid search index (FAISS + BM25) with GPU acceleration if available
  7. Registers the MCP server with Claude Code
python -m venv venv
source venv/bin/activate  # Linux/macOS
venv\Scripts\activate     # Windows

pip install git+https://github.com/RobThePCGuy/Claude-Patent-Creator.git && patent-creator setup

If you go this route, remember to activate the venv before running any manual commands. Claude Code handles activation automatically.

claude --plugin-dir ./Claude-Patent-Creator

This loads the plugin directly from your local checkout without installing from the marketplace.


CLI Commands

patent-creator setup             # Full setup wizard (downloads, builds index, registers MCP)
patent-creator health            # System health check (shows what's working and what isn't)
patent-creator status            # Same as health
patent-creator verify-config     # Check Claude Code MCP configuration
patent-creator serve             # Run the MCP server manually
patent-creator rebuild-index     # Rebuild the MPEP search index
patent-creator download-mpep     # Download MPEP PDFs only
patent-creator download-all      # Download all sources (MPEP + 35 USC + 37 CFR)
patent-creator check-bigquery    # Test BigQuery connection
patent-creator download-patents  # Download PatentsView corpus (9.2M+ patents)
patent-creator build-patent-index # Build patent search index
patent-creator patents-status    # Show patent corpus status

MCP Tools Reference

Search

ToolWhat it does
search_mpepHybrid RAG search across MPEP, 35 USC, and 37 CFR with filters
get_mpep_sectionPull full content of a specific MPEP section
search_patents_bigquerySearch 100M+ patents by keyword
get_patent_bigqueryGet full details on a specific patent
search_patents_by_cpc_bigquerySearch by CPC classification code
search_uspto_apiSearch via the USPTO API
get_uspto_patentGet patent details from USPTO
get_recent_uspto_patentsPull recent filings
search_prior_artAutomated prior art discovery

Analysis

ToolWhat it does
review_patent_claims35 USC 112(b) compliance check (definiteness, antecedent basis, structure)
review_specification35 USC 112(a) adequacy check (written description, enablement, best mode)
check_formalitiesMPEP 608 compliance (abstract, title, drawings, required sections)

Generation

ToolWhat it does
render_diagramGenerate patent-style diagrams from Graphviz DOT code
create_flowchartBuild a flowchart from a list of steps and connections
create_block_diagramBuild a block diagram from components and relationships
add_diagram_referencesAdd patent reference numbers to an existing SVG diagram
get_diagram_templatesList available diagram templates

System

ToolWhat it does
get_index_statsSearch index statistics
check_bigquery_statusBigQuery configuration status
check_diagram_tools_statusGraphviz availability
check_patent_corpus_statusPatent corpus status
check_uspto_api_statusUSPTO API connectivity
get_patent_detailsCombined patent retrieval across sources
setup_claude_configCopy .claude configuration (skills, commands) to a project directory

Skills, Agents, and Slash Commands

Skills (activate automatically)

You don't need to call these directly. Just describe what you want to do and the right skill kicks in.

SkillWhen it activatesWhat it brings
setup-assistantInstalling, configuring, or troubleshootingFull setup lifecycle guidance
patent-reviewerReviewing a complete application for complianceComprehensive review (claims + spec + formalities)
patent-claims-analyzerReviewing claims specifically for 35 USC 112(b)Deep-dive claims analysis (definiteness, antecedent basis, structure)
patent-searchSearching patents or prior artBigQuery + PatentsView API search workflows
bigquery-patent-searchQuick BigQuery-only patent searchKeyword, CPC, and patent detail retrieval across 100M+ patents
mpep-searchFinding MPEP sections or regulationsHybrid RAG search
patent-diagram-generatorCreating technical diagramsFlowcharts, block diagrams, system architectures via Graphviz
patent-application-creatorDrafting a patent application interactivelyGuided end-to-end workflow (prior art, claims, spec, diagrams, compliance)
prior-art-searchNovelty or freedom-to-operate analysis7-step prior art discovery methodology
index-managerBuilding or rebuilding the search indexMPEP index lifecycle management
development-assistantAdding features or creating toolsDevelopment workflows and patterns
troubleshooting-assistantSomething's brokenSystematic 6-step diagnostics
testing-assistantRunning tests or validationTest suite execution

Agents (long-running, work independently)

These run in the background while you keep working on other things.

AgentWhat it doesHow longOutput
patent-creatorDrafts a complete USPTO-ready application55-80 minSpecification, claims, abstract, diagrams, validation report
prior-art-searcherComprehensive prior art search15-30 minPatentability report, top 10 prior art, claim strategy, IDS list

To use them: "Create a patent for [your invention], use the patent-creator agent"

Slash Commands

/create-patent          # Complete patent creation workflow (55-80 min)
/search-prior-art       # Prior art search with novelty analysis
/full-review            # Parallel review (claims + spec + formalities)
/review-claims          # Claims-only 35 USC 112(b) analysis
/review-specification   # Specification-only 35 USC 112(a) analysis
/review-formalities     # MPEP 608 formalities check

Configuration

Environment Variables

Create a .env file in the project root (see .env.example):

# Required for BigQuery patent search
GOOGLE_CLOUD_PROJECT=your-project-id

# Optional: EPO OPS API (for EP patent full-text search)
# Free registration at https://developers.epo.org
EPO_OPS_KEY=your-consumer-key
EPO_OPS_SECRET=your-consumer-secret

# Optional API keys (for HyDE query expansion)
ANTHROPIC_API_KEY=sk-ant-...
OPENAI_API_KEY=sk-...

# Optional settings
PATENT_LOG_LEVEL=INFO
PATENT_LOG_FORMAT=human
PATENT_ENABLE_METRICS=true
PATENT_MPEP_USE_HYDE=false
PATENT_MPEP_DEVICE=gpu
PATENT_OPERATION_TIMEOUT=300

BigQuery gives you access to 100M+ worldwide patents. It requires a Google Cloud project with billing enabled (the public patent dataset itself is free to query within BigQuery's free tier).

# 1. Install Google Cloud SDK
# https://cloud.google.com/sdk/docs/install

# 2. Authenticate
gcloud auth application-default login

# 3. Set quota project (replace with your project ID)
gcloud auth application-default set-quota-project your-project-id

# 4. Set the environment variable
export GOOGLE_CLOUD_PROJECT="your-project-id"   # Linux/macOS
$env:GOOGLE_CLOUD_PROJECT="your-project-id"     # Windows PowerShell

# 5. Test it
patent-creator check-bigquery

Git Bash is required for the claude mcp add command on Windows. Install Git for Windows and set the path:

# In your .env file
CLAUDE_CODE_GIT_BASH_PATH=C:\Program Files\Git\bin\bash.exe

Use forward slashes in MCP config paths. The setup wizard handles this, but if you're configuring manually:

# Correct
clau

...

ユーザーレビュー (0)

レビューを書く

効果
使いやすさ
ドキュメント
互換性

レビューなし

統計データ

インストール数1.0K
評価4.3 / 5.0
バージョン
更新日2026年5月23日
比較事例1 件

ユーザー評価

4.3(20)
5
10%
4
40%
3
40%
2
10%
1
0%

この Skill を評価

0.0

対応プラットフォーム

🔧Claude Code

タイムライン

作成2026年4月7日
最終更新2026年5月23日