Keyword Research logo

Keyword Research

Community
liangdabiao
keyword-research

亚马逊关键词深度调研与智能分类分析。基于 Sorftime MCP 数据采集 2000+ 关键词,通过 LLM Agent 按 8 维度智能分类(否定词、品牌词、材质词、场景词、属性词、功能词、核心词、其他),生成 Markdown 报告、CSV 词库和 HTML 仪表板。触发方式:/keyword-research {ASIN} {SITE}

Overview

Publisherliangdabiao
Repositoryamazon-sorftime-research-MCP-skill
Skill namekeyword-research
Stars
964
Forks
138
Bundled files
15
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 15 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by liangdabiao on GitHub. Read the source before you install it.

Installation

Install the Keyword Research AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/liangdabiao/amazon-sorftime-research-MCP-skill/tree/main/SKILLS/skills/keyword-research
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/liangdabiao/amazon-sorftime-research-MCP-skill.git /tmp/amazon-sorftime-research-MCP-skill
mkdir -p .claude/skills
cp -r /tmp/amazon-sorftime-research-MCP-skill/SKILLS/skills/keyword-research .claude/skills/keyword-research
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Keyword Research in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Keyword Research on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Keyword Research is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

关键词调研分析 Skill

快速参考

步骤Sorftime API用途数据量
1product_traffic_terms产品流量关键词50-200
2competitor_product_keywords竞品布局关键词100-500
3category_keywords类目核心关键词100-500
4keyword_related_words长尾词扩展1000-2000
5LLM Agent8 维智能分类全量

一键执行:

bash
# 在 Claude Code 环境中运行(自动触发 LLM 分类)
python .claude/skills/keyword-research/scripts/workflow.py B07PWTJ4H1 US --claude-code-env

其他选项:

bash
# 跳过分类,仅采集数据(后续可手动LLM分类)
python .claude/skills/keyword-research/scripts/workflow.py B07PWTJ4H1 US --skip-classification

# 禁用LLM分类,使用规则分类
python .claude/skills/keyword-research/scripts/workflow.py B07PWTJ4H1 US --disable-llm-classification

触发条件

当用户使用以下方式请求时启动此分析流程:

  • 命令: /keyword-research {ASIN} {站点}
  • 示例: /keyword-research B07PWTJ4H1 US
  • 自然语言: "分析这个产品的关键词词库"、"调研 B07PWTJ4H1 的关键词"

角色设定

你是一位拥有 10 年经验的"亚马逊 PPC 广告专家"和"关键词策略分析师"。你精通亚马逊 A9 算法和关键词布局策略,能够从海量关键词中识别出高价值词和需要排除的词。


数据采集策略:方案 A(基于 ASIN 的深度分析)

输入: ASIN + 站点 + (可选) 产品信息
  ↓
┌─────────────────────────────────────────────────────────────┐
│ Step 1: 基础数据采集                                         │
├─────────────────────────────────────────────────────────────┤
│ 1. product_traffic_terms      → 产品流量词 (50-200个)       │
│ 2. competitor_product_keywords → 竞品布局词 (100-500个)     │
│ 3. category_keywords           → 类目核心词 (100-500个)      │
└─────────────────────────────────────────────────────────────┘
  ↓
┌─────────────────────────────────────────────────────────────┐
│ Step 2: 长尾词扩展                                          │
├─────────────────────────────────────────────────────────────┤
│ 从基础词中选择 Top 30 核心词                                 │
│ → 对每个调用 keyword_related_words (50-100个延伸词)         │
│ → 预计获取 1000-2000 个长尾词                               │
└─────────────────────────────────────────────────────────────┘
  ↓
┌─────────────────────────────────────────────────────────────┐
│ Step 3: 数据清洗                                            │
├─────────────────────────────────────────────────────────────┤
│ 1. 去重(归一化:小写、去除特殊字符)                        │
│ 2. 过滤无效词(过短、非英文、乱码)                          │
│ 3. 合并搜索量/CPC 等指标                                    │
└─────────────────────────────────────────────────────────────┘
  ↓
最终词库: 2000+ 关键词

关键词分类:8 维智能分类模型

分类维度

维度标识识别规则应用策略
否定/敏感词NEGATIVE与产品不相关、描述不符的词直接添加为否定关键词
品牌词BRAND竞品品牌名称竞品打法或否定
材质词MATERIAL产品材质相关词精准词组匹配
场景词SCENARIO使用场景/位置词按场景拆分广告组
属性修饰词ATTRIBUTE产品属性/特性词长尾精准匹配
功能词FUNCTION产品功能相关词广泛匹配扩流
核心产品词CORE产品核心名称大词投放占领坑位
其他OTHER未分类、拼写错误、其他语言补充埋词

分类识别示例(以 Coat Rack 为例)

产品信息: Coat Rack Wall Mount, Wood, 5 Hooks, Entryway

否定词: freestanding, over door, floor, tree, shoe
品牌词: umbra, simplehuman, mDesign, household essentials
材质词: wood, wooden, metal, aluminum, bamboo
场景词: entryway, bathroom, mudroom, garage, bedroom
属性词: wall mount, heavy duty, rustic, vintage, expandable, 5 hook
功能词: hanging, storage, organizer, display
核心词: coat rack, hook, hanger, hat rack, towel rack
其他: coatrac (拼写错误), perchero (西语)

执行流程

阶段一:数据采集

Step 1.1: 获取产品流量词
bash
curl -s -X POST "https://mcp.sorftime.com?key={API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"product_traffic_terms","arguments":{"amzSite":"US","asin":"ASIN"}}}'

返回数据: 关键词列表,包含搜索量、CPC 等指标

Step 1.2: 获取竞品布局词
bash
curl -s -X POST "https://mcp.sorftime.com?key={API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"competitor_product_keywords","arguments":{"amzSite":"US","asin":"ASIN"}}}'

返回数据: 竞品在各关键词下的排名位置

Step 1.3: 获取类目核心词
bash
# 首先获取产品详情以获取 NodeID
curl -s -X POST "https://mcp.sorftime.com?key={API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"product_detail","arguments":{"amzSite":"US","asin":"ASIN"}}}'

# 然后获取类目关键词
curl -s -X POST "https://mcp.sorftime.com?key={API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":4,"method":"tools/call","params":{"name":"category_keywords","arguments":{"amzSite":"US","nodeId":"NODE_ID"}}}'
Step 1.4: 长尾词扩展

从基础词中选择 Top 30 核心,对每个调用:

bash
curl -s -X POST "https://mcp.sorftime.com?key={API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":N,"method":"tools/call","params":{"name":"keyword_related_words","arguments":{"amzSite":"US","searchKeyword":"KEYWORD"}}}'

阶段二:LLM 智能分类

分类提示词模板
你是一位亚马逊关键词分类专家。请根据以下产品信息,将关键词列表按 8 个维度分类。

【产品信息】
产品名称: {product_name}
材质: {material}
核心属性: {features}
使用场景: {use_cases}
否定特征: {negative_features}

【分类维度】
1. NEGATIVE: 不相关的词,需直接否定
2. BRAND: 竞品品牌名称
3. MATERIAL: 材质相关词 (wood, metal, aluminum...)
4. SCENARIO: 使用场景词 (entryway, bathroom...)
5. ATTRIBUTE: 属性修饰词 (wall mount, heavy duty...)
6. FUNCTION: 功能词 (hanging, storage...)
7. CORE: 核心产品词 (coat rack, hook...)
8. OTHER: 其他(拼写错误、其他语言等)

【待分类关键词】
{keywords_json}

【输出格式】
请以 JSON 格式输出:
{
  "NEGATIVE": ["word1", "word2", ...],
  "BRAND": [...],
  ...
}
批量处理策略
  • 批次大小: 每批 150 个关键词
  • 并行处理: 可并发多个批次
  • 结果合并: 统计各分类数量,汇总关键词

阶段三:报告生成

输出文件结构
keyword-reports/
└── {ASIN}_{Site}_{YYYYMMDD}/
    ├── report.md                    # Markdown 分析报告
    ├── keywords.csv                 # 完整关键词词库(分类后)
    ├── negative_words.txt           # 否定词清单
    ├── brand_words.txt              # 品牌词清单
    ├── categorized_summary.json     # 分类统计
    └── dashboard.html               # HTML 可视化仪表板
keywords.csv 格式
csv
keyword,category,search_volume,cpc,competition,application,relevance_score
coat rack,CORE,54000,1.85,high,广泛匹配,1.00
wooden coat rack,MATERIAL,12000,1.25,medium,精准匹配,0.95
freestanding coat rack,NEGATIVE,4500,0.85,low,直接否定,0.00
...

产品信息支持(可选)

为提高分类准确性,支持用户提供产品信息:

输入方式

方式 1: 命令行参数

bash
python workflow.py B07PWTJ4H1 US --product-info product.json

方式 2: 交互式收集

bash
请输入产品核心属性(用逗号分隔):
> Wall Mount, 5 Hooks, 16.5 inches, Heavy Duty

产品信息 JSON 格式

json
{
  "product_name": "Coat Rack Wall Mount",
  "material": "Wood",
  "features": ["Wall Mount", "5 Hooks", "16.5 inches", "Heavy Duty"],
  "use_cases": ["Entryway", "Bathroom", "Mudroom", "Garage"],
  "negative_features": ["Freestanding", "Over Door", "Floor"]
}

Sorftime API 参考

关键词相关接口

接口调用消耗参数返回
product_traffic_terms1asin, site产品流量词
competitor_product_keywords1asin, site竞品布局词
category_keywords1nodeId, site类目核心词
keyword_related_words1searchKeyword, site延伸长尾词
keyword_detail1keyword, site关键词详情
product_detail1asin, site产品详情

调用格式

bash
curl -s -X POST "https://mcp.sorftime.com?key={API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":N,"method":"tools/call","params":{"name":"TOOL_NAME","arguments":{"amzSite":"US","KEY":"VALUE"}}}'

支持的站点

Amazon: US, GB, DE, FR, IN, CA, JP, ES, IT, MX, AE, AU, BR, SA


注意事项

  1. API Key 配置: 自动从 .mcp.json 读取
  2. 数据去重: 归一化处理(小写、去除特殊字符)
  3. LLM 分类: 批量处理,每批 150 个关键词
  4. 输出编码: UTF-8,支持中文和特殊字符
  5. 报告命名: {ASIN}_{Site}_{YYYYMMDD} 格式

故障排查

问题 1: API 返回 "未查询到对应产品"

原因: ASIN 不存在于 Sorftime 数据库 解决:

  1. 使用 product_search 验证 ASIN
  2. 检查站点是否正确

问题 2: 分类结果不准确

原因: 缺少产品信息上下文或使用了规则分类 解决:

  1. 提供产品信息 JSON 文件
  2. 在 Claude Code 环境中运行以使用 LLM 分类
  3. 手动执行 LLM 分类后保存到 categorized_result.json

问题 3: 长尾词扩展数量不足

原因: 核心词选择不准确或 API 限流 解决:

  1. 调整核心词选择策略,增加搜索量权重
  2. 降低 --long-tail-limit 数量避免 API 限流
  3. 使用 --skip-long-tail 跳过长尾扩展

问题 4: 分类显示 "分类失败或未提供结果" 或 使用了规则分类

原因: 在命令行环境中运行,没有触发 LLM 分类 解决:

  1. 使用 --claude-code-env 参数强制启用 LLM 分类模式:
    bash
    python workflow.py B07PWTJ4H1 US --claude-code-env
  2. 系统会输出分类提示词,复制提示词发送给 Claude 执行分类
  3. 将分类结果保存为 categorized_result.json
  4. 重新运行 workflow.py 会自动加载分类结果并重新生成报告

问题 5: 如何手动进行 LLM 分类

场景: 采集了数据但分类不准确 解决:

  1. 查看输出目录中的 classification_prompt.txt
  2. 将提示词发送给 Claude 执行分类
  3. 将分类结果保存为 categorized_result.json
  4. 运行报告重新生成脚本

参考文档


手动 LLM 分类流程

如果规则分类结果不准确,可以手动执行 LLM 分类:

步骤 1: 查看分类提示词

bash
cat keyword-reports/{ASIN}_{Site}_{YYYYMMDD}/classification_prompt.txt

步骤 2: 将提示词发送给 Claude

复制整个提示词内容,发送给 Claude 执行分类

步骤 3: 保存分类结果

将 Claude 返回的 JSON 保存到:

bash
keyword-reports/{ASIN}_{Site}_{YYYYMMDD}/categorized_result.json

步骤 4: 重新生成报告

regenerate_reports.py 支持多种使用方式:

方式 1: 从报告目录内运行(自动检测)
bash
cd keyword-reports/{ASIN}_{Site}_{YYYYMMDD}/
python ../../.claude/skills/keyword-research/scripts/regenerate_reports.py
方式 2: 指定 ASIN 和站点
bash
python .claude/skills/keyword-research/scripts/regenerate_reports.py --asin B0FG6QG8C8 --site US
方式 3: 指定完整输出目录
bash
python .claude/skills/keyword-research/scripts/regenerate_reports.py --dir "keyword-reports\B0FG6QG8C8_US_20260314"
方式 4: 列出所有可用的报告目录
bash
python .claude/skills/keyword-research/scripts/regenerate_reports.py --list

版本更新记录

v1.3 (2026-03-14)

  • ✅ 优化 regenerate_reports.py: 移除硬编码 ASIN
  • ✅ 支持多种使用方式:
    • 从报告目录内运行(自动检测)
    • 使用 --asis 和 --site 参数指定
    • 使用 --dir 参数指定完整目录
    • 使用 --list 列出所有可用报告
  • ✅ 改进错误提示: 更友好的错误信息和帮助文档

v1.2 (2026-03-14)

  • ✅ 修复 LLM 分类触发问题
  • ✅ 添加 --claude-code-env 参数强制启用 LLM 分类
  • ✅ 改进环境检测逻辑
  • ✅ 在 Claude Code 环境中自动触发 LLM 分类
  • ✅ 分类完成后自动使用 LLM 结果重新生成报告

v1.1 (2026-03-14)

  • ✅ 添加 Claude Code 环境检测
  • ✅ 添加 --skip-classification 选项
  • ✅ 改进规则分类:扩展 IP 品牌词识别
  • ✅ 改进规则分类:添加主题属性词
  • ✅ 更新故障排查文档

v1.0 (2026-03-13)

  • 初始版本
  • 支持 Sorftime API 数据采集
  • 支持 8 维智能分类
  • 生成 Markdown/CSV/HTML 报告

本技能版本: v1.3 | 最后更新: 2026-03-14

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Keyword Research AI skill do?

亚马逊关键词深度调研与智能分类分析。基于 Sorftime MCP 数据采集 2000+ 关键词,通过 LLM Agent 按 8 维度智能分类(否定词、品牌词、材质词、场景词、属性词、功能词、核心词、其他),生成 Markdown 报告、CSV 词库和 HTML 仪表板。触发方式:/keyword-research {ASIN} {SITE}

Why use Keyword Research on TypingMind?

Because you install it once and use it with any model. Keyword Research is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Keyword Research in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/liangdabiao/amazon-sorftime-research-MCP-skill/tree/main/SKILLS/skills/keyword-research. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Keyword Research?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Keyword Research?

As many as you like. As long as a model supports skills, you can use Keyword Research with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Keyword Research AI skill free?

It is published on GitHub by liangdabiao. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇