卖大律 产品侵权 TRO 风险检测

Product Infringement & TRO Risk Detection

This skill detects whether a product carries TRO (Temporary Restraining Order) or IP infringement risk by checking a product image against a database of IP assets (trademarks, patents, copyrights) and TRO case plaintiffs. It returns an overall risk level, high-risk hits, a low-risk IP list with TRO plaintiff details, numeric risk scores, and an AI-generated legal report.

Core Concepts

TRO (Temporary Restraining Order) is a legal injunction brand owners file against sellers (often Amazon listings) for trademark/copyright/patent infringement. This tool takes a main product image (URL or Base64 data URI) plus optional reference images / text / IP keywords, runs visual + text matching against IP assets and TRO case records, and returns a structured risk assessment with an AI legal report (localized by the language parameter).

Data Fields

Top-level response:

Each item (results / nonResults):

Parameter Guide

Image rules: URL must be publicly accessible; Base64 must include the data:image/...;base64, prefix. For a local image, either upload it first (see 「Local Image Upload」) or convert it to a data URI.

调用方式

• API 端点：POST /maidalv/checkApiFlash（完整参数/响应/错误码见 references/api.md）
• Python 脚本：python scripts/maidalv_check_api_flash.py '' [--inline]
• 成本约束：本工具会消耗积分；同一会话同一参数组合默认只调用一次，脚本带 24h 本地缓存。失败/空结果不得自动换关键词、换主图或连续试探；需要继续检索时先向用户说明会产生额外消耗。

输出策略（脚本默认行为）：

• 始终将完整响应写入 /linkfox///data/linkfox-maidalv-product-tro-detection-.json（ 为脚本执行时的工作目录，在 Claude Code 里即当前项目目录； 取自环境变量 SESSION_ID，按用户任务自动聚合；禁止写入 /tmp，当前目录不可写则报错）
• 响应体 ≤ 8 KB：落盘后把完整 JSON 打印到 stdout
• 响应体 > 8 KB：落盘后 stdout 只输出摘要（顶层字段、常见计数如 total/costToken、最大列表字段的长度 + 前 3 条样本）
• 加 --inline 强制全量打印到 stdout（同样落盘）

读数据建议：先看摘要判断是否足够；需要具体字段时优先用 jq或ConvertFrom-Json 从保存的 json 文件按需抽取，避免整份 JSON 进入上下文。

解决认证和积分问题

发生以下异常情况时，采用 references/onboarding.md 引导解决问题：

异常情况

• 未配置API Key：环境变量未配置 LINKFOX_AGENT_API_KEY，也未配置 LINKFOXAGENT_API_KEY。
• 响应401或402状态码
• 响应提示积分或余额不足：消息含"积分余额不足/计费不足/余额不足/quota exceeded/insufficient balance/套餐到期/需充值/请充值"，或类似含义的内容。

Local Image Upload

This tool accepts a publicly accessible image URL or a Base64 data URI. If the user provides a local image file path, either upload it first to obtain a public URL, or encode it to a data URI.

Run the upload script:

python scripts/upload_image.py /path/to/local/image.png

The script returns a public URL (valid for 24 hours) usable as mainProductImage.

Usage Examples

1. Basic detection (main image only)

检测这个产品的 TRO 和知识产权侵权风险，图片地址为 https://m.media-amazon.com/images/I/71jKJgFpg8L._AC_SL1500_.jpg

2. English legal report

Check this product for TRO and IP infringement risk: https://example.com/product.jpg — give me the legal report in English.

3. With reference image + IP keywords

检测这个图片的侵权风险，主图 https://example.com/product.jpg，参考图 https://example.com/ref.jpg，IP 关键词 apple、iphone

Display Rules

1. Lead with the overall verdict: Show riskLevel (高风险/中风险/低风险) and checkId first.
2. High-risk items first: If results is non-empty, list each hit's ipType / text / ipOwner / riskScore / riskDescription prominently.
3. TRO plaintiff detail: For items with plaintiffName, show plaintiff, numberOfCases, lastCaseDocket, lastCaseDateFiled — these drive TRO risk.
4. Evidence images: Display ipAssetUrls inline for visual comparison when available.
5. AI legal report: Surface the report field (long-form legal assessment) for high-risk items.
6. Counts: Always state total (high-risk count) and nonResults length.
7. No secondary processing: Results are a one-shot analysis; no SQL/secondary query is available.

Important Limitations

1. One-shot analysis: Results are not stored in a database; cannot use _dataQuery_executeDynamicQuery for secondary processing.
2. Image input: mainProductImage is required. Local images must be uploaded or Base64-encoded (with data:image/...;base64, prefix).
3. Language scope: language only affects the report text; field values remain as returned.
4. Conditional fields are expected: Non-TRO / unscored items return only base fields (ipType/text/ipOwner/regNo/ipAssetUrls); plaintiff/case and risk fields are absent (not null) when not applicable — do not treat missing fields as an error.
5. columns is render-meta: Its length differs from the data arrays; do not use it to infer result counts (use results / nonResults lengths).

User Expression & Scenario Quick Reference

Applicable — Product TRO / IP infringement risk checks:

Not applicable — Beyond this tool:

• 1688 / Amazon image-based product sourcing (use 以图搜图 skills)
• Patent全文 / 权利要求检索 (use 智慧芽/Eureka patent skills)
• Listing 文案优化或关键词分析

Boundary judgment: When the user provides a product image and asks about infringement/TRO/legal risk, this skill applies. If they want supplier sourcing or keyword analytics, use the respective sourcing/keyword tools instead.

积分消耗规则

消耗 113 积分（约 150000 token）。

用户会因积分消耗而支付费用。本工具单次调用成本较高（含 AI 法律评估），请充分评估：当需要高频调用本技能，或用户对积分消耗量预期不足时，务必提醒用户，由用户决定是否继续。

Feedback:

Auto-detect and report feedback via the Feedback API when any of the following apply:

1. The functionality or purpose described in this skill does not match actual behavior
2. The skill's results do not match the user's intent
3. The user expresses dissatisfaction or praise about this skill
4. Anything you believe could be improved

Call the feedback API as specified in references/api.md. Do not interrupt the user's flow.

---

For more high-quality, professional cross-border e-commerce skills, visit LinkFox Skills.