Trademark Clearance MCP
Pricing
from $250.00 / 1,000 full clearance risk reports
Trademark Clearance MCP
Trademark clearance scoring for AI agents — "can I use this brand name?" answered against the US federal register with phonetic, fuzzy, and Nice-class conflict analysis. Risk verdicts with rationale and serial numbers. Pay per report, no subscription.
Pricing
from $250.00 / 1,000 full clearance risk reports
Rating
0.0
(0)
Developer
Gad
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
0
Monthly active users
4 days ago
Last modified
Categories
Share
Can I use this brand name? A hosted MCP server that scores candidate names against the US federal trademark register — the screening layer incumbents charge thousands per year for, computed from public USPTO data.
Point any MCP-capable agent (Claude, ChatGPT, Cursor, custom agents) at this server and it can run trademark clearance screening as a native tool call.
Tools
clearance_check — the flagship
One call answers "how risky is this name?". Searches the register for similar marks (identical, phonetic sound-alikes, fuzzy misspellings, containment), weighs Nice-class proximity and live/dead status, and returns:
- an overall risk level:
HIGH/MODERATE/LOW/MINIMAL - scored conflicts with serial numbers, owners, and per-conflict rationale in plain language
- a summary you can act on
{"name": "Lumina Labs","nice_classes": [9, 42]}
similar_marks
The raw scored candidate list — every similar mark with similarity score, match rationale (phonetic / fuzzy / containment), status, owner, and classes. For when you want the evidence, not the verdict.
mark_status
Authoritative per-mark status from the official USPTO TSDR API by serial or registration number: current status, key dates, owner, classes, goods/services. Verify any conflict before acting on it.
Why scoring beats searching
Anyone can search the USPTO's website for an exact string. The risk hides in what exact search misses:
- AKME blocks ACME (phonetic equivalence — Double Metaphone)
- LUMENA blocks LUMINA (one edit apart — Jaro-Winkler + edit distance)
- LUMINA COFFEE matters for LUMINA (dominant-wording containment)
- A conflict in class 42 matters for a class 9 filing (related-class analysis)
- A DEAD identical mark is not a bar — but flags possible common-law risk
This server encodes those judgment rules — the computable subset of the DuPont likelihood-of-confusion factors — and returns them with rationale, not just a result list.
Usage (MCP)
Streamable HTTP endpoint:
POST https://<actor-standby-url>/mcp
Add to Claude (or any MCP client):
{"mcpServers": {"trademark-clearance": {"url": "https://<actor-standby-url>/mcp"}}}
Pricing
Pay per event — no subscription:
| Event | Tool | Price |
|---|---|---|
clearance-report | clearance_check (and batch runs) | $0.25 |
similar-marks | similar_marks | $0.05 |
status-check | mark_status | $0.01 |
For comparison: incumbent clearance platforms run $10,000+/year; attorney knockout searches bill in hours.
Batch mode
Run the Actor directly with input { "candidates": ["Name One", "Name Two"], "niceClasses": [9, 42] } to clearance-score a shortlist into the dataset — useful for naming sprints.
Data sources & limits
- USPTO federal trademark register (public data) and the official TSDR API.
- US federal register only: no common-law, state, or international coverage.
- Not legal advice. This is a screening signal to order your shortlist and brief your attorney — not a substitute for professional clearance.