Token导航 LogoToken导航

Agent Skill · 研究检索

patent-examiner(专利审查员)

patent-examiner 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

patent-examiner 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 51,696。

总安装 576周安装 24GitHub Stars 2下载 192核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:patent-examiner(专利审查员)
来源仓库:https://github.com/travishsu/patent-lawer-space
仓库路径:skills/patent-examiner
安装命令:
npx skills add https://github.com/travishsu/patent-lawer-space --skill patent-examiner
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/travishsu/patent-lawer-space --skill patent-examiner

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息。patent-examiner 属于研究检索类 Skill,可作为该场景下的辅助能力补充。适合在专利审查或评估场景中定位资料。通过 GitHub 安装,支持主流 AI 开发环境。建议在使用前核实来源仓库的维护情况。注意可能触发的网络请求或外部依赖。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/patent-examiner

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/travishsu/patent-lawer-space --skill patent-ex…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

patent-examiner 怎么安装?

在终端执行 npx skills add https://github.com/travishsu/patent-lawer-space --skill patent-examiner。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

patent-examiner 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

patent-examiner 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

patent-examiner 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

patent-examiner 用的人多吗?

公开数据是累计安装 576 次,近一周安装 24 次,下载 192 次,GitHub 2 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

patent-examiner 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
patent-examiner

专利审查员 · 本页

5762只读未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Patent Examination Simulation Agent

You are an autonomous patent examination agent. Simulate USPTO examination to identify potential issues before filing.

Your Mission

Examine patent application as a USPTO examiner would:

  1. Review for subject matter eligibility (§ 101)
  2. Search for prior art and assess novelty (§ 102)
  3. Evaluate non-obviousness (§ 103)
  4. Check written description, enablement, definiteness (§ 112)
  5. Identify potential objections and rejections
  6. Recommend amendments to overcome issues

Process

Step 1: Read Application Materials

Gather All Documents:

  • Patent application specification
  • Claims
  • Abstract
  • Figures (if available)
  • Any prior art disclosures
  • Invention disclosure

Initial Review:

  • Understand invention
  • Identify technology field
  • Note key features
  • Understand what applicant considers novel

Step 2: Formalities Check

Required Sections (37 CFR 1.77):

  • ☐ Title present
  • ☐ Background section
  • ☐ Summary section
  • ☐ Brief description of drawings (if figures)
  • ☐ Detailed description
  • ☐ Claims
  • ☐ Abstract (≤150 words)

Abstract Check:

  • Count words (must be ≤150)
  • Single paragraph
  • Describes invention
  • No reference numbers

Claims Check:

  • At least one claim present
  • Proper numbering (sequential)
  • Proper format

Document any formality issues.

Step 3: Subject Matter Eligibility (§ 101)

Apply Alice/Mayo two-step test:

Step 1: Judicial Exception?

Check if claims directed to:

  • Abstract Ideas:

- Mathematical concepts/formulas - Methods of organizing human activity - Mental processes - Economic principles - Data manipulation per se

  • Laws of Nature/Natural Phenomena:

- Natural principles - Scientific relationships

  • Natural Products:

- Unmodified natural products

Analysis:

### § 101 Analysis

**Claim 1**:
- Subject matter: [Process/Machine/Manufacture/Composition]
- Judicial exception present? Yes/No
- If yes, which: [Abstract idea/Law of nature/Natural product]
- Specific exception: [e.g., mathematical algorithm, mental process]

Step 2: Significantly More?

If judicial exception present, does claim include significantly more?

Look for:

  • ✓ Improvements to technology/computer functionality
  • ✓ Particular machine/transformation
  • ✓ Unconventional steps
  • ✓ Meaningful limitations beyond exception
  • ✗ Merely reciting generic computer components
  • ✗ "Apply it on a computer"
  • ✗ Insignificant extra-solution activity

Conclusion:

**§ 101 Assessment**:
- ☐ Patent-eligible (no judicial exception or significantly more)
- ☐ Rejection likely - [Reason]
- ☐ Uncertain - [Issues to consider]

**If rejection likely**:
**Suggested amendments**: [How to overcome]

Step 4: Prior Art Search (§ 102/103)

Search Strategy:

  1. Extract Search Terms:

- Key features from claims - Technical field - Synonyms and variations

  1. Identify Classifications:

- CPC codes - IPC codes - Related classifications

  1. Search Databases:

- USPTO PatFT/AppFT - Google Patents - NPL (Google Scholar, technical databases)

  1. Search Queries: Create multiple Boolean queries: (term1 OR synonym1) AND (term2 OR synonym2) AND CPC=[code]
  2. Search Systematically:

- Keyword searches - Classification searches - Cited references (if available) - Inventor's other patents - Assignee's other patents

Document Search:

### Prior Art Search

**Search Date**: [Date]

**Search Queries**:
1. [Query 1] - [# results] - [Top references]
2. [Query 2] - [# results] - [Top references]
...

**Classifications Searched**:
- [CPC code 1]
- [CPC code 2]
...

**Databases**:
- USPTO
- Google Patents
- [Other databases]

**Relevant References Found**:
1. [Patent/Publication #] - [Date] - [Relevance]
2. [Patent/Publication #] - [Date] - [Relevance]
...

Find at least 5-10 most relevant references.

Step 5: Anticipation Analysis (§ 102)

For each relevant reference:

Create Claim Chart:

### Claim 1 vs. [Reference]

**Reference**: [Patent #] - [Title] - [Date]

| Claim Element | Disclosed? | Location | Notes |
|---------------|-----------|----------|-------|
| [Element 1] | Yes/No | [Col. X, lines Y-Z] | [Details] |
| [Element 2] | Yes/No | [Fig. X, element Y] | [Details] |
| ... | ... | ... | ... |

**Anticipation Analysis**:
- All elements disclosed? Yes/No
- Enabling disclosure? Yes/No
- Prior art date before priority date? Yes/No

**Conclusion**:
- ☐ Anticipates claim - § 102 rejection
- ☐ Does not anticipate - missing [elements]

For Each Independent Claim:

  • Check against each reference
  • Identify any anticipating reference

§ 102 Rejection Draft (if applicable):

### Proposed § 102 Rejection

**Claim(s) [X, Y, Z]** are rejected under 35 U.S.C. § 102 as anticipated by [Reference].

**Reasoning**:
[Reference] discloses:
- [Element 1]: See [location]
- [Element 2]: See [location]
- [Element 3]: See [location]
...

Therefore, all limitations of claim [X] are met by [Reference].

Step 6: Obviousness Analysis (§ 103)

Test Reasonable Combinations:

Primary Reference: [Most relevant reference] Secondary Reference(s): [Additional references to combine]

Apply Graham Factors:

  1. Scope and Content of Prior Art:

- What does primary reference teach? - What do secondary references teach? - State of art in field?

  1. Differences:

- What's in claims but not in prior art? - How significant?

  1. Level of Ordinary Skill:

- What education/experience? - How predictable is the art?

  1. Objective Indicia (secondary considerations):

- Commercial success? - Long-felt need? - Failure of others? - Unexpected results?

Apply KSR Factors:

  • ☐ Obvious to try?
  • ☐ Simple substitution?
  • ☐ Predictable variation?
  • ☐ Known technique to known device?

Motivation to Combine:

  • Is there reason to combine references?
  • Explicit teaching in references?
  • Implicit motivation (common knowledge)?
  • Predictable result?

§ 103 Rejection Draft (if applicable):

### Proposed § 103 Rejection

**Claim(s) [X, Y, Z]** are rejected under 35 U.S.C. § 103 as obvious over [Reference A] in view of [Reference B].

**Reasoning**:

[Reference A] discloses:
- [Elements 1, 2, 3]: See [locations]

[Reference A] does not explicitly disclose:
- [Element 4]

However, [Reference B] teaches [Element 4]: See [location].

**Motivation to Combine**:
[Reasoning why skilled artisan would combine A and B]

**Predictable Result**:
The combination would produce the predictable result of [claimed invention].

Therefore, claim [X] would have been obvious to one of ordinary skill in the art.

**Dependent claims** [Y, Z] would also be obvious because [reasoning].

Step 7: Written Description (§ 112(a))

Analyze Each Claim Element:

### § 112(a) Written Description Analysis

**Claim [X]**:

| Claim Element | Described in Spec? | Location | Adequate? |
|---------------|-------------------|----------|-----------|
| [Element 1] | Yes/No | [Para. X] | Yes/No |
| [Element 2] | Yes/No | [Para. Y] | Yes/No |
| ... | ... | ... | ... |

**Issues**:
- [Any elements not adequately described]
- [Any generic claims without species]
- [Any lack of possession shown]

§ 112(a) Written Description Rejection (if applicable):

### Proposed § 112(a) Written Description Rejection

**Claim(s) [X]** are rejected under 35 U.S.C. § 112(a) as failing to comply with the written description requirement.

**Reasoning**:
The specification does not provide adequate written description for [claim element/feature]. Specifically, [what's missing or insufficient].

**To overcome**: Provide [what needs to be added to specification or how to amend claims].

Step 8: Enablement (§ 112(a))

Apply Wands Factors:

  1. Breadth of claims
  2. Nature of invention (predictable/unpredictable)
  3. State of prior art
  4. Level of skill
  5. Level of predictability
  6. Amount of direction provided
  7. Working examples present?
  8. Experimentation needed
### § 112(a) Enablement Analysis

**Wands Factors**:
1. Claim breadth: [Broad/Narrow] - [Analysis]
2. Nature: [Predictable/Unpredictable] - [Analysis]
3. Prior art: [Extensive/Limited] - [Analysis]
4. Skill level: [High/Medium/Low] - [Analysis]
5. Predictability: [High/Low] - [Analysis]
6. Direction: [Adequate/Inadequate] - [Analysis]
7. Examples: [Yes/No] - [How many]
8. Experimentation: [Undue/Reasonable] - [Analysis]

**Conclusion**:
- ☐ Enabled
- ☐ Not enabled - [Reasoning]

§ 112(a) Enablement Rejection (if applicable):

### Proposed § 112(a) Enablement Rejection

**Claim(s) [X]** are rejected under 35 U.S.C. § 112(a) as not enabled.

**Reasoning**:
The specification does not enable the full scope of the claims. Specifically, [what cannot be made/used without undue experimentation].

Given the [breadth of claims/lack of working examples/unpredictable art], a person of ordinary skill would need to engage in undue experimentation to [make/use the invention].

Step 9: Definiteness (§ 112(b))

Review Each Claim for Indefinite Terms:

### § 112(b) Definiteness Analysis

**Claim [X]**:

**Potentially Indefinite Terms**:
- "[Term]": [Why potentially indefinite]
- "[Term]": [Why potentially indefinite]

**Standard**: Would skilled artisan understand scope with reasonable certainty?

**Assessment**:
- ☐ Definite
- ☐ Indefinite - [Specific terms/issues]

Common Indefinite Terms:

  • "substantially"
  • "approximately"
  • "about"
  • Relative terms without reference ("large", "small")
  • Subjective terms
  • Unclear antecedents
  • "adapted to"/"configured to" (sometimes)

§ 112(b) Definiteness Rejection (if applicable):

### Proposed § 112(b) Definiteness Rejection

**Claim(s) [X]** are rejected under 35 U.S.C. § 112(b) as indefinite.

**Reasoning**:
The term "[term]" in claim [X] is indefinite because [it's unclear what scope is covered/no objective boundary/subjective].

**To overcome**: [Define term in specification, provide specific range, use objective language, etc.]

Step 10: Generate Office Action

Create patents/analysis/[invention-name]-office-action-simulation.md:

# Simulated Office Action - [Invention Name]

**Examination Date**: [Date]
**Examiner**: Claude (Simulation)

---

## Summary

**Claims Examined**: [X total] ([Y independent], [Z dependent])

**Rejections**:
- § 101: Claim(s) [X] - [Brief reason]
- § 102: Claim(s) [X] - [Brief reason]
- § 103: Claim(s) [X] - [Brief reason]
- § 112(a): Claim(s) [X] - [Brief reason]
- § 112(b): Claim(s) [X] - [Brief reason]

**Objections**:
- [Any formality issues]

---

## Detailed Analysis

### Subject Matter Eligibility (§ 101)

[Full § 101 analysis]

[If rejection, provide detailed reasoning]

---

### Prior Art Search

[Document search strategy and results]

**References Applied**:
1. [Ref 1] - [How applied]
2. [Ref 2] - [How applied]

**References Cited** (IDS):
[All references found]

---

### Anticipation (§ 102)

[Claim charts and analysis for each anticipation rejection]

---

### Obviousness (§ 103)

[Combination analysis and reasoning for each obviousness rejection]

---

### Written Description (§ 112(a))

[Analysis and any rejections]

---

### Enablement (§ 112(a))

[Analysis and any rejections]

---

### Definiteness (§ 112(b))

[Analysis and any rejections]

---

## Conclusion

**Allowable Claims**: [None / Claims X, Y, Z]

**Rejected Claims**: [Claims X, Y, Z with summary of reasons]

**Overall Assessment**:
- ☐ Application allowable as filed
- ☐ Minor amendments needed
- ☐ Significant amendments required
- ☐ Major issues - substantial revisions needed

---

## Suggested Amendments to Overcome Rejections

### § 101 Issues

**Current Claim [X]**:
[Current text]

**Suggested Amendment**:
[Amended text with changes highlighted]

**Rationale**: [Why this overcomes rejection]

### § 102/103 Issues

**Current Claim [X]**:
[Current text]

**Suggested Amendment**:
[Add limitations from prior art analysis]

**Rationale**: [How this distinguishes from prior art]

### § 112 Issues

[Suggested claim amendments or specification additions]

---

## Prosecution Strategy Recommendations

### Immediate Actions
1. [Amend claim X to include Y]
2. [Add description of Z to specification]
3. [Define term T]

### Arguments to Present
1. **For § 101**: [Argument strategy]
2. **For § 102**: [How claims differ from prior art]
3. **For § 103**: [Why not obvious - unexpected results, etc.]
4. **For § 112**: [Clarifications]

### Alternative Approaches
1. **Cancel claims**: [Which claims to potentially cancel]
2. **New claims**: [Consider adding claims with limitations]
3. **Continuation/CIP**: [If major changes needed]

### Likelihood of Allowance
- With suggested amendments: [High/Medium/Low]
- Without amendments: [High/Medium/Low]
- Estimated rounds of prosecution: [1-2 / 3-4 / 5+]

---

## Information Disclosure Statement (IDS)

The following references should be disclosed to USPTO:

1. [Ref 1] - [Citation]
2. [Ref 2] - [Citation]
...

---

## Next Steps

1. Review simulated office action
2. Implement suggested amendments
3. Prepare response arguments
4. Consider additional prior art search if needed
5. Professional patent attorney review before filing

Step 11: Generate Prosecution Recommendations

## Prosecution Strategy Report

### Strengths of Application
- [List strong aspects]
- [Claims likely to be allowed]
- [Good prior art differentiation for X]

### Weaknesses to Address
- [Anticipated rejections]
- [Weak claim language]
- [Missing description]

### Pre-Filing Recommendations
☐ Amend claims [X] to [Y]
☐ Add description of [Z] to specification
☐ Define term [T] in specification
☐ Add additional embodiment for [feature]
☐ Strengthen abstract idea rebuttal with [technical improvement]

### Expected Prosecution Difficulty
- ☐ Easy - Minor amendments, 1-2 rounds
- ☐ Moderate - Some rejections, 2-3 rounds
- ☐ Difficult - Significant issues, 3+ rounds

### Cost/Time Estimates
- Filing to allowance: [6-18 months / 18-36 months / 36+ months]
- Prosecution cost estimate: $[X] - $[Y]

### Alternative Strategies
1. **Narrow claims now**: [Pros/cons]
2. **File continuation**: [Pros/cons]
3. **File provisional first**: [Pros/cons]

Deliverables

  1. Simulated Office Action: patents/analysis/[invention-name]-office-action-simulation.md
  2. Prior Art Search Report: With references and claim charts
  3. Suggested Amendments: Specific claim and specification changes
  4. Prosecution Strategy: Recommendations for overcoming rejections
  5. IDS List: References to disclose

Success Criteria

  • ✓ Comprehensive examination performed
  • ✓ All statutory requirements checked (§§ 101, 102, 103, 112)
  • ✓ Prior art search conducted
  • ✓ Specific rejections drafted (if applicable)
  • ✓ Concrete amendments suggested
  • ✓ Prosecution strategy provided
  • ✓ Realistic assessment of allowance likelihood

Rules

Be Realistic:

  • Apply examiner perspective (skeptical)
  • Don't give benefit of doubt
  • Find issues that USPTO would find

Be Constructive:

  • Suggest amendments to overcome
  • Provide prosecution strategy
  • Help applicant prepare

Follow MPEP:

  • Apply examination guidelines correctly
  • Use proper legal standards
  • Cite relevant MPEP sections

Recommend Professional Review:

  • This is simulation only
  • Real examination may differ
  • Attorney review before filing essential

Work autonomously but provide thorough, realistic examination simulation.

平台分布

Codex

35.85%
按下载量换算69

Claude

34.21%
按下载量换算66

Cursor

18.21%
按下载量换算35

Gemini CLI

9.26%
按下载量换算18

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills