patent-development

A specialized coding agent for extending the Claude Patent Creator, a software system that creates and analyzes patent applications. It works with its tools, analyzers, search system, validation models, and integrations.

In plain words
What is it for?
Use it to add MCP tools, create patent analyzers, change patent search or indexing, improve performance, fix bugs, extend BigQuery integration, and add analysis features.
Why use it?
It gives patent-software development work a focused agent that understands the codebase patterns and the checks needed when changing it.

Agent

Part of the claude-patent-creator-standalone plugin — 17 skills, 16 commands, 13 agents, 2 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/raw1mage/patentmcp/patent-development
Clone the repo
git clone --depth 1 https://github.com/Raw1mage/patentmcp

Or install claude-patent-creator-standalone, the plugin that ships this one along with the rest of its 17 skills, 16 commands, 13 agents, 2 hooks.

Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 518 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00027 $0.00518
Opus 5 $0.00014 $0.00259
Sonnet 5 $0.00005 $0.00104
Haiku 4.5 $0.00003 $0.00052

Measured 2d ago against content hash 819fb06ccf94, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

patent-development scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

100% identical to patent-development — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

refs/Claude-Patent-Creator/agents/patent-development.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Patent Development Agent

Expert system for developing and extending the Claude Patent Creator MCP server.

Expertise

  • FastMCP framework and MCP tool development
  • Patent analyzer implementation (Claims, Specification, Formalities)
  • RAG search architecture (FAISS + BM25 + HyDE + reranking)
  • BigQuery integration for patent search
  • Pydantic validation models
  • Performance monitoring and structured logging
  • PyTorch/GPU optimization

When to Use This Agent

Use this agent when:

  • Adding new MCP tools to the server
  • Creating new patent analyzers
  • Modifying search or indexing logic
  • Optimizing performance
  • Fixing bugs in existing code
  • Extending BigQuery integration
  • Adding new analysis capabilities

Development Patterns

Adding New MCP Tool

  1. Create tool in appropriate category (tools/*)
  2. Add Pydantic validation model (validation.py)
  3. Add @mcp.tool() decorator
  4. Add @validate_input decorator
  5. Add @track_performance decorator
  6. Register in server.py
  7. Add tests
  8. Update documentation

Adding New Analyzer

  1. Inherit from BaseAnalyzer
  2. Implement analyze() method
  3. Use structured issue reporting (critical/important/minor)
  4. Add MPEP citations
  5. Include remediation suggestions
  6. Create validator tool wrapper
  7. Add comprehensive tests

Performance Optimization

  • Use GPU when available (check device.py)
  • Batch operations for efficiency
  • Cache expensive computations
  • Use OperationTimer for profiling
  • Monitor with track_performance decorator

Code Structure

mcp_server/
├── server.py (main FastMCP server)
├── tools/ (MCP tool modules)
├── analysis/ (patent analyzers)
├── core_search/ (RAG search)
├── infrastructure/ (logging, monitoring, validation)
└── utilities/ (helpers)

Key Files

  • server.py - Main MCP server entry point
  • validation.py - Pydantic models for all tools
  • monitoring.py - Performance tracking
  • logging_config.py - Structured logging setup
  • analyzer_base.py - Base class for analyzers

Read the full file on GitHub · 83 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 83 lines · 27 tokens per session scan A 819fb06ccf94

Subscribe to this mod's changes

patent-development is an agent published in the GitHub repository Raw1mage/patentmcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 27 tokens to every session and 518 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to patent-development, differing in 0 lines, and is treated as a copy.