Multi-Jurisdiction Clearance Workflow

IPKit provides normalized trademark search results across 10 jurisdictions through a single interface. It integrates directly with Claude Desktop or ChatGPT, giving you clearance search, Nice class analysis, goods and services drafting, and filing readiness assessment without switching between individual trademark office websites.

This guide walks through a complete clearance workflow suitable for professional use.

Installation

Hosted Mode (Recommended Start)

The fastest way to start is using IPKit’s hosted server. No credentials or local setup required.

Add to your Claude Desktop configuration:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "ipkit": {
      "command": "npx",
      "args": [
        "-y", "mcp-remote",
        "https://ipkit.fly.dev/mcp",
        "--header", "Authorization:Bearer YOUR_API_KEY"
      ]
    }
  }
}

Restart Claude Desktop after saving.

Local Mode (Full Control)

For higher rate limits and direct control over provider credentials, run IPKit locally. This requires API credentials from the trademark offices you want to search.

{
  "mcpServers": {
    "ipkit": {
      "command": "node",
      "args": ["/path/to/core/dist/index.js"],
      "env": {
        "USPTO_API_KEY": "your_key",
        "EUIPO_CLIENT_ID": "your_id",
        "EUIPO_CLIENT_SECRET": "your_secret",
        "IPAUSTRALIA_CLIENT_ID": "your_id",
        "IPAUSTRALIA_CLIENT_SECRET": "your_secret",
        "EPO_CONSUMER_KEY": "your_key",
        "EPO_CONSUMER_SECRET": "your_secret"
      }
    }
  }
}

See the Configuration Reference for all available environment variables and how to obtain credentials from each office.

Phase 1: Nice Class Selection

Before searching, establish the correct Nice classes for your client’s goods and services. Misclassified searches produce both false positives and false negatives.

Suggest classes from a business description

“Suggest Nice classes for a company that provides cloud-based legal practice management software with document automation and client billing.”

The suggest_nice_classes tool analyzes the description and recommends classes with explanations. For the example above, you would likely see classes 9 (software), 35 (business management), 36 (financial services), 42 (SaaS/technology services), and possibly 45 (legal services).

Verify class scope

Confirm what each recommended class covers:

“Look up Nice class 42 and show me the full description with examples.”

The nice_class_lookup tool returns the official class heading, common goods/services, and related classes. It also identifies product-service class pairs (e.g., class 9 software products often pair with class 42 software services).

Phase 2: Clearance Search

Primary clearance analysis

Run the comprehensive clearance tool with your target classes and jurisdictions:

“Run trademark clearance for ‘LexFlow’ in classes 9, 35, and 42 across the US, EU, and AU.”

The trademark_clearance tool executes a multi-layered search:

1. Exact and substring matching against each jurisdiction’s database
2. Truncated stem search for marks longer than 4 characters (catches prefix-similar marks that substring matching misses)
3. Famous marks pre-check against approximately 200 well-known marks (catches misspellings that registry search misses, e.g., “GOOGEL” matching “GOOGLE”)
4. Ensemble similarity scoring combining Jaro-Winkler distance, n-gram overlap, Damerau-Levenshtein distance, and jurisdiction-aware phonetic algorithms

Each conflict is scored and assessed with an overall risk level (low/medium/high/critical) and specific recommendations.

Supplementary targeted searches

For conflicts that require closer examination, or to broaden coverage beyond the clearance tool’s automated scope:

“Search for trademarks owned by ‘LexFlow Technologies’ in the US and EU.”

The trademark_search tool supports multiple search types:

Search Type	Use Case
name	Find marks containing a specific string
fuzzy	Similarity search with ensemble scoring (threshold 0.7)
owner	Search by applicant/owner name
number	Look up a specific application or registration number

Jurisdiction-aware phonetic matching

IPKit uses different phonetic algorithms depending on the jurisdiction:

Jurisdiction	Algorithm	Why
US, GB, CA	NYSIIS	Optimized for English name patterns
EU	Cologne Phonetic	Handles German/continental European consonant patterns
AU, NZ	Caverphone	Regional pronunciation patterns
JP, CN	Disabled	Not applicable to CJK characters
WIPO	Double Metaphone	General-purpose cross-linguistic fallback

This means the same clearance search produces different phonetic similarity scores depending on the jurisdiction, reflecting how examiners in each office would assess aural similarity.

Phase 3: Deep-Dive on Conflicts

For each material conflict flagged by the clearance analysis, pull full details:

“Get the full status and history for trademark US-97123456.”

The trademark_status tool returns:

• Current registration status and all lifecycle dates
• Complete goods and services specification
• Owner/applicant details and representative information
• Status history (where available)
• Priority claims and convention details

This is the data you need to assess whether a conflict is genuinely blocking or distinguishable.

Phase 4: Goods and Services Specification

Generate specifications

Once classes are confirmed and clearance looks favorable, draft the filing specification:

“Generate a goods and services specification for a cloud-based legal practice management platform with document automation, client billing, and calendar integration. Target the US and EU, classes 9 and 42.”

The generate_gs_specification tool produces jurisdiction-specific specification text:

• US (USPTO): TEAS Plus/Standard compatible language
• EU (EUIPO): TMclass compatible terms
• WIPO (Madrid Protocol): Madrid-compatible specifications

Validate against EUIPO Harmonised Database

For EU filings, verify that your terms are pre-approved:

“Validate these G&S terms against the EUIPO Harmonised Database: ‘Computer software for legal practice management’ in class 9.”

The validate_gs_terms tool checks each term against the HDB and reports:

• Whether the term is harmonized (pre-approved by all EU offices for faster examination)
• Any errors (invalid characters, forbidden terms, syntax issues)
• Any warnings (duplicated terms, length recommendations)

Translate for multi-language filings

EUIPO applications require filing in two languages. Madrid Protocol filings may need multiple translations:

“Translate ‘Computer software for legal practice management’ from English to French and German.”

The translate_gs_terms tool uses HDB-harmonized translations across 23 EU official languages, ensuring the translated terms are also pre-approved.

Phase 5: Filing Readiness Assessment

Use the built-in clearance prompt to generate a structured filing readiness summary:

“Assess filing readiness for ‘LexFlow’ in classes 9 and 42 for the US and EU markets.”

This produces a structured assessment covering:

• Clearance search summary with risk level
• Distinctiveness evaluation
• Recommended filing strategy (direct national vs. Madrid Protocol)
• Identified conflicts requiring attorney judgment
• Suggested next steps

Multi-Jurisdiction Strategy

Different markets call for different search strategies:

Market	Primary Search	Why
United States	US + WIPO	USPTO database plus Madrid designations targeting the US
European Union	EU + WIPO	EUIPO covers all 27 member states; WIPO covers Madrid designations
Australia / NZ	AU + NZ + WIPO	Trans-Tasman markets often filed together
United Kingdom	GB + EU + WIPO	Post-Brexit, GB is separate from EU; many EU marks have UK equivalents
Global launch	ALL	Search all 9 jurisdictions simultaneously

Tip

When searching “ALL” jurisdictions, IPKit queries all configured offices in parallel. If a provider fails (e.g., rate limit or temporary outage), the results include partial data with per-jurisdiction error information in the metadata. Check metadata.errors in the response.

Monitoring and Portfolio Management

After filing, set up ongoing monitoring:

“Create a watch for status changes on trademark EU-018901234.”

“Create a similar-filing watch for ‘LexFlow’ in classes 9 and 42 across the US and EU.”

The monitoring system tracks status transitions, new conflicting filings, and approaching deadlines (opposition periods, renewals, expiry). Webhook delivery is available for external integrations. See the Monitoring guide for details.

Limitations and Disclaimers

What IPKit provides:

• Normalized search results from 10 trademark offices
• Algorithmic similarity analysis (phonetic, visual, conceptual)
• Nice class recommendations and G&S specification drafting
• Structured risk assessments with scoring

What IPKit does not replace:

• Professional judgment on likelihood of confusion
• Common law trademark searches (unregistered rights, state registrations, domain names)
• Logo and design mark similarity analysis (word marks only)
• Official legal opinions or attorney-client privilege
• Direct filing with trademark offices

Data currency: Results come from live queries against trademark office APIs. However, most offices have a publication lag of days to weeks between filing and database availability. Very recent filings may not appear in search results.

Jurisdiction coverage: US, EU, AU, NZ, and WIPO provide full search capabilities. GB, CA, JP, and CN are limited to number-based lookups only (no name or owner search).