Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Draft patent claims for an invention. Use when user says "撰写权利要求", "draft claims", "写权利要求书", "claim drafting", or wants to create patent claims. The core skill of the patent pipeline.
.claude/skills/wanshuiyin-claims-drafting/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 149% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 142% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 165% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 72% | 0% |
Draft patent claims based on: $ARGUMENTS
This is the most critical skill in the patent pipeline. Claims define the legal scope of protection -- everything else (specification, figures, abstract) exists to support and enable the claims.
REVIEWER_MODEL = gpt-6-astra — External examiner for claim quality reviewMAX_CLAIM_REVISION_ROUNDS = 3 — Maximum revision iterationsCLAIM_STYLE = "auto" — US (Jepson or open), EP (two-part mandatory), CN (two-part), auto (detect from jurisdiction)MIN_INDEPENDENT_CLAIMS = 2 — Typically method + system. For utility model (实用新型): apparatus/device only, NO method claims.MAX_TOTAL_CLAIMS = 20 — Practical limit (USPTO includes 20 in base fee)PATENT_TYPE = "invention" — invention (发明专利) or utility_model (实用新型, apparatus claims only)patent/INVENTION_DISCLOSURE.md — structured invention with core/supporting/optional featurespatent/PRIOR_ART_REPORT.md — prior art to avoidpatent/NOVELTY_ASSESSMENT.md — novelty analysis with suggested amendments$ARGUMENTSLoad ../shared-references/patent-writing-principles.md for claim drafting principles, antecedent basis rules, and common pitfalls. Load ../shared-references/patent-format-cn.md for CN claim format (其特征在于). Load ../shared-references/patent-format-us.md for US claim format (comprising, means-plus-function). Load ../shared-references/patent-format-ep.md for EP two-part form (characterised in that).
Based on patent type and jurisdiction:
If PATENT_TYPE = utility_model (实用新型):
MIN_INDEPENDENT_CLAIMS = 1 (single apparatus claim is sufficient)Based on target jurisdiction:
| Jurisdiction | Claim Style | Characterising Phrase | Preamble Format | |-------------|------------|----------------------|-----------------| | CN | Two-part (两部式) | 其特征在于 | 一种...的方法/装置,包括: | | US | Open (preferred) | comprising | A method for..., comprising: | | EP | Two-part (mandatory) | characterised in that | A method for..., comprising known], characterised in that inventive] | | ALL | Draft CN + US + EP | All of the above | All of the above |
CRITICAL — Claims numbering (CN format):
CRITICAL — No empirical content in claims:
For each claim category identified in INVENTION_DISCLOSURE.md:
Method Claim (broadest):
System/Apparatus Claim:
Quality checks for each independent claim:
For each independent claim, draft 5-10 dependent claims that:
Dependent claim format:
Rules for dependent claims:
Create a preliminary mapping to verify enablement:
| Claim Element | Must be described in specification | Reference numeral | |---------------|-----------------------------------|-------------------| | element 1] | Yes/No | numeral] | | element 2] | Yes/No | numeral] |
If any element lacks specification support, add it to the specification requirements.
Start the examiner review with a dedicated Codex reviewer agent at xhigh reasoning:
spawn_agent:
model: gpt-6-astra
reasoning_effort: xhigh
message: |
You are a senior patent examiner at the [USPTO/CNIPA/EPO].
Review the following patent claims for quality and patentability.
CLAIMS: [all claims]
PRIOR ART: [prior art references from PRIOR_ART_REPORT.md]
INVENTION: [summary from INVENTION_DISCLOSURE.md]
Analyze each claim for:
1. Clarity (35 USC 112(b) / Art 84 EPC): Are terms definite?
2. Written description support: Does the spec support all claim scope?
3. Anticipation (102/Art 54): Would any single reference anticipate?
4. Obviousness (103/Art 56): Would any combination render obvious?
5. Claim scope: Are independent claims broad enough to be valuable?
6. Dependent claims: Do they provide meaningful fallback positions?
7. Antecedent basis: Any issues with "a"/"the" usage?
8. Indefinite terms: Any functional/result language issues?
For each issue found, provide:
- The specific claim number and element
- The problem (cite statute/rule)
- A suggested fix
Provide an overall PATENTABILITY SCORE: 1-10.If the examiner review identifies issues:
send_input using the saved reviewer idMAX_CLAIM_REVISION_ROUNDS timestextsend_input: target: [saved reviewer id from Step 5] message: | Here is the revised claim set after addressing the previous office action. Revised claims: [full revised claims] Focus only on unresolved or newly introduced issues.
Write patent/CLAIMS.md:
markdown## Patent Claims ### Independent Claims #### Claim 1 — Method [formatted claim text] #### Claim X — System/Apparatus [formatted claim text] ### Dependent Claims #### Claim 2 (depends on 1) [formatted claim text] #### Claim 3 (depends on 1) [formatted claim text] ... ### Claims Summary Table | Claim | Type | Depends On | Key Limitation | Prior Art Avoidance | |-------|------|-----------|----------------|---------------------| | 1 | Method | — | [core inventive features] | [what makes it novel over prior art] | | 2 | Method | 1 | [narrowing] | [additional distinguishing] | | X | System | — | [mirrors claim 1] | [same as claim 1] | ... ### Examiner Review Summary [Key findings and how they were addressed]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-14 | pass→pass | 15,094 | 14,164 | -6% | 1 | 1 | 0% | 1,659 | 3,964 | +139% | 0 | 0 | — |
case-01 | fail→fail | 15,730 | 16,772 | +7% | 1 | 1 | 0% | 237 | 2,831 | +1095% | 0 | 0 | — |
case-02 | fail→fail | 17,485 | 14,801 | -15% | 1 | 1 | 0% | 421 | 2,777 | +560% | 0 | 0 | — |
case-03 | fail→fail | 21,443 | 15,708 | -27% | 1 | 1 | 0% | 369 | 2,939 | +696% | 0 | 0 | — |
case-04 | fail→fail | 52,743 | 13,952 | -74% | 1 | 1 | 0% | 8,234 | 2,791 | -66% | 0 | 0 | — |
case-05 | fail→fail | 26,697 | 28,637 | +7% | 1 | 1 | 0% | 3,420 | 6,493 | +90% | 0 | 0 | — |
case-06 | pass→pass | 16,262 | 19,717 | +21% | 1 | 1 | 0% | 1,844 | 4,999 | +171% | 0 | 0 | — |
case-07 | pass→fail | 19,278 | 17,029 | -12% | 1 | 1 | 0% | 2,311 | 3,084 | +33% | 0 | 0 | — |
case-08 | fail→pass | 17,794 | 20,248 | +14% | 1 | 1 | 0% | 2,107 | 5,243 | +149% | 0 | 0 | — |
case-09 | fail→pass | 19,796 | 24,199 | +22% | 1 | 1 | 0% | 2,364 | 5,712 | +142% | 0 | 0 | — |
case-10 | fail→pass | 18,465 | 7,815 | -58% | 1 | 1 | 0% | 2,053 | 2,913 | +42% | 0 | 0 | — |
case-11 | fail→pass | 18,252 | 9,171 | -50% | 1 | 1 | 0% | 1,213 | 3,212 | +165% | 0 | 0 | — |
case-12 | fail→pass | 21,193 | 17,915 | -15% | 1 | 1 | 0% | 2,608 | 4,479 | +72% | 0 | 0 | — |
case-13 | pass→pass | 20,760 | 18,853 | -9% | 1 | 1 | 0% | 2,398 | 4,670 | +95% | 0 | 0 | — |
case-15 | pass→pass | 20,029 | 16,354 | -18% | 1 | 1 | 0% | 2,185 | 4,186 | +92% | 0 | 0 | — |
case-16 | pass→pass | 16,687 | 14,515 | -13% | 1 | 1 | 0% | 1,827 | 4,039 | +121% | 0 | 0 | — |
case-17 | fail→pass | 17,545 | 17,108 | -2% | 1 | 1 | 0% | 1,962 | 4,580 | +133% | 0 | 0 | — |
case-18 | pass→pass | 12,733 | 10,659 | -16% | 1 | 1 | 0% | 1,235 | 3,338 | +170% | 0 | 0 | — |
case-19 | pass→pass | 15,196 | 12,861 | -15% | 1 | 1 | 0% | 1,495 | 3,786 | +153% | 0 | 0 | — |
case-20 | pass→pass | 14,266 | 15,163 | +6% | 1 | 1 | 0% | 1,617 | 4,242 | +162% | 0 | 0 | — |
case-21 | fail→fail | 15,947 | 8,220 | -48% | 1 | 1 | 0% | 1,585 | 3,089 | +95% | 0 | 0 | — |
case-22 | pass→pass | 18,976 | 20,554 | +8% | 1 | 1 | 0% | 2,258 | 4,976 | +120% | 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. 22 cases were attempted, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 17 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/11/2026 | +27% |
Other measured skills in the registry, with their headline benchmark lift.