Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.

MITAuto-check passedLegal & Compliance

Install Mpep Search

skills CLI
$ npx skills add RobThePCGuy/Claude-Patent-Creator --skill mpep-search -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install RobThePCGuy/Claude-Patent-Creator mpep-search --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/RobThePCGuy/Claude-Patent-Creator.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/mpep-search .claude/skills/mpep-search && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
mpep-search
GitHub stars
196
Token cost
~978 tokens
SKILL.md length
255 words
Files
2
Skills in repo
17
Repo updated
First seen
Licence
MIT

At a glance

Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.

  • Works in 2 steps: search_mpep → get_mpep_section
  • Tasks that involve Legal research
  • SKILL.md covers Core Operations, Input Validation and Implementation Notes
  • Runs Python scripts from its folder

What it does

Mpep Search is an agent skill from RobThePCGuy/Claude-Patent-Creator. Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.

Its SKILL.md is about 980 tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `mpep_search.py`).

It sits in Legal & Compliance, covering Legal research, Intellectual property and Vector databases. It works with Model Context Protocol. The repository describes itself as: USPTO patent creation system with MCP server + Claude Code plugin. Hybrid RAG search over MPEP/USC/CFR, BigQuery access to 76M+ patents, automated 35 USC 112 compliance checks… The licence is MIT.

When your agent uses it

  • Tasks that involve Legal research
  • Tasks that involve Intellectual property
  • Tasks that involve Vector databases

Example prompts

  • “/mpep-search”

Requirements

  • Python 3

Workflow steps

2 steps, taken from the step headings in SKILL.md.

  1. search_mpep
  2. get_mpep_section

What it can do on your machine

Read from SKILL.md and the folder at commit a089731. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships script files (Python), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Mpep Search loads about 978 tokens when it runs. Until then it costs about 29 tokens; SKILL.md has 255 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~29
When it runs · the whole SKILL.md, loaded when a task matches
~978

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from RobThePCGuy/Claude-Patent-Creator at commit a089731, republished under its MIT licence (© RobThePCGuy). 255 words, ~978 tokens.

Download SKILL.mdSave it as .claude/skills/mpep-search/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
mpep-search
description
Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.

MPEP Search Skill

Search MPEP corpus through hybrid RAG (FAISS vector + BM25 keyword + HyDE + cross-encoder reranking).

Sources:

  • MPEP: Manual of Patent Examining Procedure
  • 35 USC: United States Code Title 35
  • 37 CFR: Code of Federal Regulations Title 37
  • Subsequent Publications: Federal Register updates (post-Jan 2024)

Core Operations

1. search_mpep

Inputs:

  • query (string, required): Search query (minimum 3 characters)
  • top_k (int, optional): Number of results (default: 5, max: 20)
  • retrieve_k (int | None, optional): Candidates before reranking (default: top_k * 4, max: 100)
  • source_filter (string | None, optional): Filter by source ("MPEP", "35_USC", "37_CFR", "SUBSEQUENT", or None)
  • is_statute (bool | None, optional): Filter for statute content
  • is_regulation (bool | None, optional): Filter for regulation content
  • is_update (bool | None, optional): Filter for recent updates

Outputs:

python
{
    "rank": int,
    "source": str,
    "section": str,
    "file": str,
    "page": int,
    "has_statute": bool,
    "has_mpep_ref": bool,
    "has_rule_ref": bool,
    "is_statute": bool,
    "is_regulation": bool,
    "is_update": bool,
    "relevance_score": float,
    "text": str,
    # Optional for SUBSEQUENT:
    "doc_type": str,
    "fr_citation": str,
    "effective_date": str
}

Examples:

python
# Basic search
search_mpep("enablement requirement 35 USC 112", top_k=5)

# Search only statutes
search_mpep("written description", top_k=10, is_statute=True)

# Search recent updates
search_mpep("AI inventorship", is_update=True)

# Filter by source
search_mpep("fee schedule", source_filter="37_CFR")
2. get_mpep_section

Retrieve all content from specific MPEP section.

Inputs:

  • section_number (string, required): MPEP section number (e.g., "2100", "608.01")
  • max_chunks (int, optional): Maximum chunks to return (default: 50)

Outputs:

python
{
    "section": str,
    "total_chunks": int,
    "chunks": [
        {
            "text": str,
            "metadata": {
                "source": str,
                "file": str,
                "page": int,
                "section": str,
                "has_statute": bool,
                "has_mpep_ref": bool,
                "has_rule_ref": bool,
                "is_statute": bool,
                "is_regulation": bool,
                "is_update": bool
            }
        }
    ]
}

Error Response:

python
{"error": "No content found for MPEP section {section_number}"}

Examples:

python
# Get MPEP 2100 (Patentability)
get_mpep_section("2100", max_chunks=50)

# Get subsection
get_mpep_section("608.01")

Input Validation

Query validation:

  • Minimum 3 characters
  • Case-insensitive
  • No empty/whitespace-only queries

Section number validation:

  • Numeric with optional decimal (e.g., "100", "2100", "608.01")

Limits:

  • top_k capped at 20
  • retrieve_k capped at 100

Implementation Notes

Index Location:

  • FAISS index: mcp_server/index/mpep_index.faiss
  • Metadata: mcp_server/index/mpep_metadata.json
  • BM25 index: mcp_server/index/mpep_bm25.json

Search Architecture:

  1. HyDE Query Expansion (hypothetical documents)
  2. Hybrid Retrieval (FAISS vector + BM25 keyword via RRF)
  3. Cross-Encoder Reranking (final relevance scores)
  4. Metadata Filtering (source/type filters)

Dependencies:

  • sentence-transformers (BGE-base-en-v1.5)
  • FAISS (vector search)
  • rank-bm25 (keyword search)
  • Cross-encoder (reranking)
  • HyDE (optional, graceful degradation)

Error Handling:

  • Clear error messages for missing index/invalid queries
  • Graceful degradation if HyDE fails
  • Input validation before processing

© RobThePCGuy, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/mpep-search of RobThePCGuy/Claude-Patent-Creator.

  • SKILL.md
  • mpep_search.py

Open the folder on GitHubat commit a089731

Compare with similar skills

Mpep Search next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Mpep Search compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Mpep Search this skillRobThePCGuy/Claude-Patent-Creator196—~978Automated safety check: PassMIT
Find Law Firmjeremylongshore/tons-of-skills-marketplace2.8k—~3.6kAutomated safety check: NotesMIT
Sealeap Hundun Amazon Brand Adjacent Keyword Ip Riskxjli360/sealeap-amazon-skills247—~821Automated safety check: PassMIT
Tw Legal RAGaa0101181514/tw-legal-rag328—~580Automated safety check: PassCustom licence
Patsnap Ip Searchingpatsnap/mcp113—~1.1kAutomated safety check: PassApache-2.0
USPTO Patent and Trademark Datadavila7/claude-code-templates32k11 repos~4.6kAutomated safety check: PassMIT

Similar skills

  • Find Law Firm

    jeremylongshore/tons-of-skills-marketplace

    A skill your agent uses whenever the user wants to find, shortlist, vet, or enrich US B2B law firms — corporate, IP/patent, M&A and securities, employment, commercial litigation…

    2.8k GitHub stars~3.6k tokensUpdated today
    Legal & ComplianceAuto-check: notes
  • Screen whether a candidate core keyword or brand name sits too close to an existing trademark or an unresolved utility patent before building ranking around it, and follow a time-boxed response…

    247 GitHub stars~821 tokensUpdated 11 days ago
    Legal & ComplianceAuto-check passed
  • Tw Legal RAG

    aa0101181514/tw-legal-rag

    Retrieve real Taiwan court judgments with verifiable citations before answering any question about Taiwan law or case law.

    328 GitHub stars~580 tokensUpdated today
    Legal & ComplianceAuto-check passed
  • Patsnap Patent Research MCP for AI agents. An agent skill from patsnap/mcp.

    113 GitHub stars~1.1k tokensUpdated 1 mo ago
    Legal & ComplianceAuto-check passed
  • USPTO Patent and Trademark Data

    davila7/claude-code-templates

    Searches USPTO patent and trademark data through its APIs: patent search, PEDS examination history, assignments, citations, office actions and TSDR status.

    32k GitHub starsUsed in 11 repos~4.6k tokens
    Legal & ComplianceAuto-check passed
  • Finnish Legal Document Review

    akunikkola/claude-for-legal-finland

    Checks a Finnish legal document, such as a contract, statement, appeal or decision, in eight stages, looking for errors, gaps and risks before it is signed or submitted.

    109 GitHub stars~3.4k tokensUpdated 3 mo ago
    Legal & ComplianceAuto-check passed

More from RobThePCGuy/Claude-Patent-Creator

All 17 skills in this repo
  • Bigquery Patent Search

    RobThePCGuy/Claude-Patent-Creator

    Fast, cloud-based patent searching across 100 million+ worldwide patents using Google BigQuery - keyword search, CPC classification, patent details retrieval

    196 GitHub stars~1.4k tokensUpdated 2 days ago
    Auto-check: notes
  • Development Assistant

    RobThePCGuy/Claude-Patent-Creator

    Guides through adding new features, MCP tools, analyzers, and extending the patent creator system.

    196 GitHub stars~1.3k tokensUpdated 2 days ago
    Auto-check: notes
  • Epc Search

    RobThePCGuy/Claude-Patent-Creator

    RAG search across EPC provisions, EPO Guidelines for Examination, and PCT rules for European and international patent law research

    196 GitHub stars~2.2k tokensUpdated 2 days ago
    Auto-check passed
  • Epo Patent Analyzer

    RobThePCGuy/Claude-Patent-Creator

    Automated analysis of patent applications for EPO compliance with Art.

    196 GitHub stars~2k tokensUpdated 2 days ago
    Auto-check passed
  • Epo Patent Search

    RobThePCGuy/Claude-Patent-Creator

    Search European patents using EPO OPS API (full-text, legal status, families) and BigQuery (100M+ patents, EP filter) for prior art, competitive intelligence, and freedom-to-operate analysis

    196 GitHub stars~1.9k tokensUpdated 2 days ago
    Auto-check passed
  • Index Manager

    RobThePCGuy/Claude-Patent-Creator

    Manages MPEP index lifecycle including downloads, building, maintenance, and optimization.

    196 GitHub stars~994 tokensUpdated 2 days ago
    Auto-check passed

Questions about Mpep Search

What does Mpep Search do?

Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates. Mpep Search is an agent skill from RobThePCGuy/Claude-Patent-Creator. Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.

When should I use Mpep Search?

Mpep Search fits situations like: tasks that involve Legal research; tasks that involve Intellectual property; tasks that involve Vector databases.

How do I install Mpep Search in Claude Code?

Run `npx skills add RobThePCGuy/Claude-Patent-Creator --skill mpep-search -a claude-code`. Or copy the skill folder (skills/mpep-search in RobThePCGuy/Claude-Patent-Creator) into .claude/skills/mpep-search in your project. Claude Code loads it when a task matches its description.

How do I install Mpep Search in Codex?

Run `npx skills add RobThePCGuy/Claude-Patent-Creator --skill mpep-search -a codex`. Or copy the skill folder (skills/mpep-search in RobThePCGuy/Claude-Patent-Creator) into .agents/skills/mpep-search in your project. Codex loads it when a task matches its description.

Can I use Mpep Search in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add RobThePCGuy/Claude-Patent-Creator --skill mpep-search -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/mpep-search, .gemini/skills/mpep-search, .github/skills/mpep-search and .opencode/skills/mpep-search in your project.

What does Mpep Search need to run?

Going by SKILL.md and its folder, Mpep Search needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Mpep Search access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Mpep Search safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Mpep Search use?

Mpep Search is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Mpep Search use?

About 978 tokens (SKILL.md is roughly 3.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Mpep Search?

Skills that share tags, products or a category with Mpep Search: Find Law Firm (jeremylongshore/tons-of-skills-marketplace, 2.8k stars), Sealeap Hundun Amazon Brand Adjacent Keyword Ip Risk (xjli360/sealeap-amazon-skills, 247 stars), Tw Legal RAG (aa0101181514/tw-legal-rag, 328 stars) and Patsnap Ip Searching (patsnap/mcp, 113 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Mpep Search?

RobThePCGuy (a GitHub user) maintains it in RobThePCGuy/Claude-Patent-Creator, which has 196 GitHub stars. The repository holds 17 skills in this directory. The repository was last updated on October 6, 2026.

Source: RobThePCGuy/Claude-Patent-Creator on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.