Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Korean patent / trademark / design search via KIPRIS Plus OpenAPI (특허청). Search patents (특허/실용신안), trademarks (상표), and designs (디자인) by keyword, applicant, or application/registration number. Returns clean JSONL for AI agents. Use for prior-art search, competitor IP monitoring, brand-name availability checks, M&A due diligence, and Korean IP analytics. Requires a free KIPRIS_PLUS_KEY from plus.kipris.or.kr.
.claude/skills/kipris-cli/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | — | — |
| case-15 | ✗→✓ | ▲ Improved | — | — |
| case-10 | ✗→✓ | ▲ Improved | — | — |
| case-09 | ✗→✓ | ▲ Improved | — | — |
| case-02 | ✗→✓ | ▲ Improved | — | — |
Five subcommands that wrap the KIPRIS Plus OpenAPI (특허청 한국특허정보원) — Korea's authoritative IP database covering ~5M patents, ~6M trademarks, ~1M designs. Every endpoint returns XML; this skill normalizes it to JSONL for AI-agent consumption.
> Why this matters: Korean IP lookups are otherwise locked behind kipris.or.kr's web UI. AI agents doing prior-art search, M&A target diligence, or brand-name availability checks need clean JSON access. This is the only ClawHub skill covering KIPRIS at the time of publish.
bashexport KIPRIS_PLUS_KEY="<your-32-char-key>" # from plus.kipris.or.kr (free, ~1d approval) # Patent search by keyword bin/kipris-cli patent --word "양자컴퓨팅" --rows 10 # Trademark availability check bin/kipris-cli trademark --word "AURORA" --rows 5 # Design search bin/kipris-cli design --word "smart watch" # Patent bibliographic detail by application number bin/kipris-cli patent-detail --app-no 1020230012345 # Search patents by applicant name (개인/법인) bin/kipris-cli applicant --name "삼성전자주식회사" --rows 20
| Command | Endpoint | Use case | |---|---|---| | patent | patUtiModInfoSearchSevice/patUtiModInfoSearch | Patent + 실용신안 keyword search | | trademark | TrademarkInfoSearchService/trademarkInfoSearchInfo | Brand availability + competitor TM watch | | design | DesignInfoSearchService/designInfoSearchInfo | Industrial design search | | patent-detail | patUtiModInfoSearchSevice/patUtiModBibliographicInfoSearch | Full bibliographic detail by 출원번호 | | applicant | patUtiModInfoSearchSevice/patUtiModInfoSearchByApplicantName | All filings by applicant (출원인) |
--word "<query>" — free-text Korean or English (UTF-8)--app-no <13-digit> — 출원번호 (1020230012345 = 특허, 4020210012345 = 상표, 3020230012345 = 디자인)--rows N — page size (default 30, max 500)--page N — 1-indexed page (default 1)--format json|xml|jsonl — output (default jsonl)--key <KEY> — override $KIPRIS_PLUS_KEYpatent)--applicant "<name>" — 출원인명--inventor "<name>" — 발명자명--ipc "<code>" — IPC 분류 (e.g. G06N)--pat true|false — include 특허 (default true)--utility true|false — include 실용신안 (default true)--last-update YYYYMMDD — 최종변경일 sincetrademark)--applicant "<name>"--class "<NICE int>" — 상품분류 (NICE classification 1-45)--reg-status "registered|pending|rejected|expired" — 권리상태--start-date YYYYMMDD / --end-date YYYYMMDD — 출원일 rangeEach line is one record (patent/TM/design). Common fields normalized across types:
json{ "type": "patent", "app_no": "1020230012345", "title_ko": "양자 회로 최적화 방법 및 장치", "title_en": "Method and apparatus for quantum circuit optimization", "applicant": "삼성전자주식회사", "inventors": ["홍길동", "김철수"], "app_date": "20230101", "reg_no": null, "reg_date": null, "pub_no": "1020240054321", "pub_date": "20240715", "ipc": ["G06N10/40", "G06F17/14"], "abstract": "..." }
Trademarks add nice_class, tm_kind (문자/도형/입체); designs add locarno_class, parts.
ServiceKey and export KIPRIS_PLUS_KEY=....> ⚠️ The legacy kipris.or.kr API endpoints have been deprecated since 2024Q4. This skill uses the current plus.kipris.or.kr/kipo-api/kipi/... paths.
examples/competitor-patent-watch.sh — daily JSONL feed of new patents filed by a competitor (paginated + dedup).examples/brand-availability-check.sh — check if a brand name has TM filings in NICE classes 9, 35, 42 (typical SaaS classes).examples/m-and-a-ip-snapshot.sh — given a 사업자등록번호 (via nts-bizno-cli) → corporate name → all KIPRIS filings → consolidated CSV. Pairs naturally with opendart-cli for full M&A due-diligence chain.opendart-cli — corporate disclosures (pair for IP-vs-financial due diligence)nts-bizno-cli — KYB business-number lookup (anchor entity name)juso-address-cli — registered address resolutionkosis-cli — R&D investment statistics overlaypatent --rows 30 --page 1 = 1 call. The skill caches nothing — caller is responsible for pagination throttling.0.05s between calls.MIT-0. No warranty. Built using only the publicly documented KIPRIS Plus OpenAPI specification.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +82 percentage points is the difference between those two pass rates over the 22 comparable cases.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.