Trademark Basic Search

Use this skill for naming-stage trademark risk screening. Treat the result as a basic search, not legal clearance.

Core Rules

• Always say this is not legal advice and not lawyer-level trademark clearance.
• Prefer official databases and official search entry points over generic web snippets.
• Distinguish "no obvious exact hit found" from "cleared"; never say a mark is legally safe.
• Search exact names, spacing variants, root-word variants, and adjacent industry terms.
• For current trademark, legal, domain, or company facts, browse or use live sources.

Workflow

1. Build the query set.
   • Exact: proposed mark, uppercase form, spaced form, common spelling variants.
   • Adjacent: roots, suffixes, obvious abbreviations, competitors, same-word-family marks.
   • Classes: use Class 9 and Class 42 for developer tools/software by default; add product-specific classes when relevant.
2. Run USPTO structured search.
   • Use scripts/uspto_tmsearch.py from this skill.
   • Keep the smoke test enabled unless there is a strong reason not to; it verifies the USPTO endpoint with APPLE.
   • Example:

python3 ~/.skills/laodao-skills/trademark-basic-search/scripts/uspto_tmsearch.py \
  PRAGMOKIT "PRAGMO KIT" PRAGMAKIT "PRAGMA KIT" PRAGMO "PRAGMATIC SEMICONDUCTOR" \
  --format markdown

3. Check non-US official sources.
   • Read references/search-sources.md when you need the current official entry points and limitations.
   • WIPO Global Brand Database may show ALTCHA protection; do not script around it.
   • TMview/EUIPO pages are front-end apps; do not treat failed automated parsing as a clean result.
   • CNIPA/China search may require registration/login; record that manual or attorney review is needed.
4. Run quoted web searches.
   • Search exact quoted terms, e.g. "PragmoKit", "PRAGMOKIT", "Pragmo Kit".
   • Search adjacent terms with product context, e.g. "Pragmo" software, "Pragmatic Semiconductor" trademark.
   • Record only relevant uses: same/similar mark, same/similar market, same/similar class.
5. Write the report.
   • Include date, mark, target market, target classes, exact queries, adjacent queries, sources, and limitations.
   • Separate confirmed structured results from manual-check requirements.
   • Include a practical conclusion: continue, hold, rename, or escalate to attorney search.

Report Template

## 基础商标检索记录

检索日期：YYYY-MM-DD
拟用名称：<brand>
检索性质：基础排雷，不是法律意见或正式 trademark clearance。

### 检索范围

- 精确名称：...
- 相邻名称：...
- 重点类别：Class 9, Class 42, ...

### USPTO 基础结果

<paste script markdown table>

### WIPO / EUIPO / TMview / CNIPA 状态

- WIPO：...
- EUIPO / TMview：...
- CNIPA：中国区需要人工或代理复核。

### 公开 web 排雷

- Exact match：...
- Adjacent risk：...

### 基础结论

<continue / hold / rename / attorney-review recommendation>

### 参考来源

- <official links>

USPTO Script Notes

The script calls the USPTO TM Search backend:

https://tmsearch.uspto.gov/prod-v1-0-0/tmsearch

It sends an Elasticsearch-style POST query using:

• term on WM
• match_phrase on WM
• query_string against wordmark and wordmarkPseudoText

If the script fails or the smoke test returns zero results, do not rely on the output. Use the official USPTO web interface manually and state the limitation.

Final Wording

Use wording like:

• "基础检索未发现明显精确命中。"
• "相邻风险来自 ..."
• "公开发布、logo 投入、商业化或重点市场申请前，需要律师级正式检索。"

Avoid wording like:

• "商标可用。"
• "已经 cleared。"
• "没有风险。"
• "可以放心注册。"