Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Patent research and IP landscape: prior-art search, claim mapping, landscape analysis, and freedom-to-operate. Use when conducting a prior-art search, mapping an IP landscape, evaluating patentability, or running an FTO analysis.
.claude/skills/borghei-patent/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 177% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 195% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 200% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 147% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 70% | 0% |
A research-focused patent skill (not legal filing). Covers prior-art search, IP landscape mapping, claim analysis, freedom-to-operate, and patentability. For formal patent prosecution, work with a licensed patent attorney.
Before running the analysis, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
prior_art_search_planner.py to produce a search plan.bashpython3 patent/scripts/prior_art_search_planner.py \ --input invention.json --format markdown
claim_landscape_mapper.py to cluster by claim type, owner,technology subarea, recency; surface white space + crowded areas.
bashpython3 patent/scripts/claim_landscape_mapper.py \ --input patents.json --format markdown
patentability_scorer.py to rate novelty, non-obviousness,utility, subject matter eligibility.
bashpython3 patent/scripts/patentability_scorer.py \ --input patentability.json --format markdown
Plus:
Inventors often miss non-patent prior art; this is where searches break.
Different from patentability. FTO asks: can I commercialize without infringing someone else's patent?
references/prior-art-search-strategy.md — search databases, classifications, query patternsreferences/claim-mapping-and-landscape.md — claim analysis, landscape visualizationreferences/freedom-to-operate-and-patentability.md — FTO process, patentability criterialegal/contract-review — IP licensing contractsc-level-advisor/general-counsel-advisor — strategic IP counselresearch/litreview — non-patent literature search overlap| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,816 | 29,109 | +22% | 1 | 1 | 0% | 3,879 | 6,370 | +64% | 0 | 0 | — |
case-02 | fail→fail | 25,353 | 37,166 | +47% | 1 | 1 | 0% | 3,838 | 7,353 | +92% | 0 | 0 | — |
case-03 | fail→fail | 21,918 | 23,541 | +7% | 1 | 1 | 0% | 3,311 | 5,309 | +60% | 0 | 0 | — |
case-04 | pass→pass | 17,503 | 19,141 | +9% | 1 | 1 | 0% | 2,704 | 4,595 | +70% | 0 | 0 | — |
case-05 | fail→pass | 4,637 | 2,370 | -49% | 1 | 1 | 0% | 696 | 1,926 | +177% | 0 | 0 | — |
case-06 | fail→pass | 9,195 | 2,294 | -75% | 1 | 1 | 0% | 654 | 1,930 | +195% | 0 | 0 | — |
case-07 | fail→pass | 4,145 | 1,921 | -54% | 1 | 1 | 0% | 620 | 1,863 | +200% | 0 | 0 | — |
case-08 | pass→pass | 14,374 | 13,512 | -6% | 1 | 1 | 0% | 2,288 | 3,707 | +62% | 0 | 0 | — |
case-09 | pass→pass | 14,362 | 12,899 | -10% | 1 | 1 | 0% | 2,309 | 3,565 | +54% | 0 | 0 | — |
case-10 | pass→pass | 12,696 | 12,964 | +2% | 1 | 1 | 0% | 1,953 | 3,484 | +78% | 0 | 0 | — |
case-11 | pass→pass | 15,482 | 20,191 | +30% | 1 | 1 | 0% | 2,534 | 4,702 | +86% | 0 | 0 | — |
case-12 | pass→pass | 14,891 | 15,777 | +6% | 1 | 1 | 0% | 2,239 | 4,010 | +79% | 0 | 0 | — |
case-13 | pass→pass | 15,191 | 14,082 | -7% | 1 | 1 | 0% | 2,425 | 3,782 | +56% | 0 | 0 | — |
case-14 | pass→pass | 19,873 | 19,885 | +0% | 1 | 1 | 0% | 2,865 | 4,727 | +65% | 0 | 0 | — |
case-15 | pass→pass | 15,408 | 13,560 | -12% | 1 | 1 | 0% | 2,270 | 3,535 | +56% | 0 | 0 | — |
case-16 | pass→pass | 14,921 | 23,869 | +60% | 1 | 1 | 0% | 2,203 | 5,153 | +134% | 0 | 0 | — |
case-17 | pass→pass | 14,285 | 15,681 | +10% | 1 | 1 | 0% | 2,080 | 3,986 | +92% | 0 | 0 | — |
case-18 | pass→pass | 17,788 | 17,761 | -0% | 1 | 1 | 0% | 2,507 | 4,208 | +68% | 0 | 0 | — |
case-19 | pass→pass | 15,072 | 19,398 | +29% | 1 | 1 | 0% | 2,571 | 4,785 | +86% | 0 | 0 | — |
case-20 | pass→pass | 12,062 | 15,794 | +31% | 1 | 1 | 0% | 1,946 | 4,074 | +109% | 0 | 0 | — |
case-21 | pass→pass | 15,404 | 17,203 | +12% | 1 | 1 | 0% | 2,688 | 4,558 | +70% | 0 | 0 | — |
case-22 | fail→pass | 6,610 | 6,338 | -4% | 1 | 1 | 0% | 1,016 | 2,514 | +147% | 0 | 0 | — |
case-23 | pass→pass | 22,756 | 31,426 | +38% | 1 | 1 | 0% | 3,989 | 7,024 | +76% | 0 | 0 | — |
case-24 | pass→pass | 13,875 | 14,783 | +7% | 1 | 1 | 0% | 2,106 | 3,687 | +75% | 0 | 0 | — |
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. 24 cases were attempted. The headline lift of +17 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.