yuc16/PatentRadar / skills/patentradar
patentradar
Verified专利侵权竞品分析 skill。输入专利公开号(CN/US/EP/JP 等),输出可人工复核的 markdown 竞品分析报告。工作流:拆解权利要求 → 搜索市场竞品 → 抓证据(含图像)→ 全部权利要求逐特征对比 → 生成报告。触发词:专利侵权分析、专利竞品、claim chart、专利公开号、CN/US/EP/JP 专利号、专利对比、技术特征对比、专利保护范围、侵权风险评估。不适用:专利申请文书撰写、专利无效宣告、知识产权法律咨询、专利诉讼策略。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 13.1 KB |
| agents/competitor_search.md | markdown | 30.9 KB |
| agents/decompose.md | markdown | 7.4 KB |
| agents/full_claim_chart.md | markdown | 20.1 KB |
| agents/report.md | markdown | 11.7 KB |
| configs/technology_tags.toml | config | 8.1 KB |
| schemas/full_claim_chart_report.md | markdown | 4.3 KB |
| schemas/full_claim_chart_report.schema.json | config | 5.1 KB |
| schemas/task_package.md | markdown | 3.3 KB |
| schemas/task_package.schema.json | config | 2.6 KB |
| schemas/top_competitor_report.md | markdown | 4.9 KB |
| schemas/top_competitor_report.schema.json | config | 4.1 KB |
| schemas/validate.py | script | 2.9 KB |
| scripts/render_pdf.py | script | 2.7 KB |
schemas/full_claim_chart_report.md
View on GitHub# full_claim_chart.json schema
模块三输出 → 模块四输入。
## 顶层结构
```json
{
"publication_no": "CN114512759B",
"top_competitors": [<FullClaimChartCandidate>, ...],
"excluded_candidates": [<FullClaimChartCandidate>, ...]
}
```
| 字段 | 类型 | 必填 | 说明 |
|---|---|---|---|
| `publication_no` | string | ✅ | 同模块二 |
| `top_competitors` | FullClaimChartCandidate[] | ✅ | 通过模块三补搜后非失效的候选 |
| `excluded_candidates` | FullClaimChartCandidate[] | ✅ | 模块二就失效或模块三新发现失效的候选 |
## FullClaimChartCandidate
```json
{
"candidate": {<同模块二 candidate>},
"launch_date": "...",
"launch_date_evidence": [...],
"disqualified": false,
"disqualification_reason": "",
"claim_charts": [<ClaimChartEntry>, ...],
"claim_1_score": 90.0,
"total_score": 90.0,
"searched_queries": [...],
"searched_providers": []
}
```
| 字段 | 类型 | 说明 |
|---|---|---|
| `candidate` | object | 同模块二 CandidateEvidence.candidate(结构完全一致)|
| `launch_date` / `launch_date_evidence` | — | 同模块二,模块三复用;若新拿到证据证明早于申请日则更新为 disqualified |
| `disqualified` / `disqualification_reason` | — | 同模块二,仅"权 1 任一明确不满足"或"launch_date 早于专利申请日"触发 |
| `claim_charts` | ClaimChartEntry[] | **全部权利要求**的对比(按 claim_no 1, 2, 3, ... 顺序)|
| `claim_1_score` | number 0-100 | 权 1 总分(= claim_charts[claim_no==1].claim_score)|
| `total_score` | number 0-100 | **只看权 1**:total_score = claim_1_score |
| `searched_queries` | string[] | 模块三跑过的 query(不包括模块二的)|
| `searched_providers` | string[] | 用过的搜索后端 |
## ClaimChartEntry
```json
{
"claim_no": 1,
"claim_text": "1.一种锂离子电池电芯...",
"comparisons": [<FeatureComparison>, ...],
"claim_score": 90.0
}
```
| 字段 | 类型 | 说明 |
|---|---|---|
| `claim_no` | int | 权利要求编号,与 task_package.claims[].claim_no 对应 |
| `claim_text` | string | 完整原文(与 task_package 一致)|
| `comparisons` | FeatureComparison[] | 该 claim 全部 feature 的判定 |
| `claim_score` | number 0-100 | `mean(该 claim 各 feature.score) × 100`;任一"明确不满足"该 claim 直接 0 |
## FeatureComparison(含 evidence_gap_brief)
```json
{
"feature_id": "C1-F3",
"patent_feature": "...",
"competitor_feature": "...",
"status": "可能满足",
"score": 0.8,
"evidence": [...],
"reasoning": "...",
"suggested_followup_queries": [],
"evidence_gap_brief": "还缺:...\n下一步建议:..."
}
```
字段同模块二,新增/强化:
| 字段 | 模块三阶段要求 |
|---|---|
| `suggested_followup_queries` | 终判后必须为空数组(已搜完,不再有 followup)|
| `evidence_gap_brief` | **权 1 中 status ∈ {可能满足, 证据不足} 的 feature 必填**(两行结构:还缺 / 下一步建议);其他情况一律 `""` |
## evidence_gap_brief 格式(仅权 1 缺口 feature 必填)
两行文字:
```
还缺:<具体技术维度,写清"已有公开证据为何不足以证明本特征,还缺 XXX 才能证明该特征">
下一步建议:<明确建议去哪里(具体网站名如畅易汽车网、汽车之家、汽修巴巴等真实网站)找什么>
```
**写作要求**:
- "还缺"具体到技术维度,不能笼统写"证据不足";同时说清"为何不足"和"还缺什么才能证明"
- "下一步建议"必须明确可执行——指定真实网站名 + 在该网站做什么动作("定位 XX 维修手册""下载 XX 规格书""查看 XX 章节")
- **不要写"搜 XXX"这种 query 字串**——只给方向
- 带 URL 必须是真实网站根域名,不能瞎编具体文章路径
- 不超过 2 行;每行不超过 120 字
## 关键约束
- `claim_charts` 必须按 `claim_no` 1, 2, 3, ... 顺序排列
- `claim_charts` 覆盖 task_package 中**全部权利要求**(独立 + 从属)
- 模块二已判 1.0 的权 1 feature **沿用判定** + 同 URL evidence(除非有新证据冲突)
- 模块三**不重复**模块二跑过的 query(应换思路)
- **失效只看权 1**:从属权利要求"明确不满足" → 对应 `claim_score = 0`,但**不**触发候选整体 `disqualified=true`
- `total_score = claim_1_score`(非权 1 不进 ranking 总分)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |