Content
# amazon sorftime research MCP skill -
Based on Sorftime MCP service and Claude Skills' Amazon competitive product analysis toolset.
## Project Introduction
This project configures the MCP (Model Context Protocol) server for Sorftime cross-border e-commerce data service and develops the `amazon-analyse` skill for comprehensive analysis of Amazon competitive product listings.
### Core Features
- **Competitive Product Listing Analysis**: Automatically retrieves product details, reviews, keywords, and trend data
- **Keyword Analysis**: Traffic sources, competitor layout, long-tail keyword mining
- **Review Sentiment Analysis**: Advantage clustering, pain point identification, improvement suggestions
- **Market Insight**: Seasonal trends, competitive landscape, opportunity identification
---
## Quick Start
### Environment Requirements
- Claude Code CLI
- Valid Sorftime API Key
- Bash shell environment
### Install MCP Server
The MCP configuration file is located at `.mcp.json`, which has configured the Sorftime service:
```json
{
"mcpServers": {
"sorftime": {
"type": "streamableHttp",
"url": "https://mcp.sorftime.com?key=YOUR_API_KEY",
"name": "Sorftime MCP"
}
}
}
```
### Use Analysis Skills
```bash
# Analyze competitive products in the US
/amazon-analyse B07PWTJ4H1 US
# Analyze competitive products in Europe
/amazon-analyse B08N5WRWNW DE
```
After the analysis is completed, the report will be automatically saved to the `reports/` directory.
---
## Sorftime MCP Service
### Supported Amazon Sites
| Site | Code | Site | Code |
|------|------|------|------|
| United States | US | Mexico | MX |
| United Kingdom | GB | United Arab Emirates | AE |
| Germany | DE | Australia | AU |
| France | FR | Brazil | BR |
| India | IN | Saudi Arabia | SA |
| Canada | CA | - | - |
| Japan | JP | - | - |
| Spain | ES | - | - |
| Italy | IT | - | - |
### Available Interfaces (25)
#### Product Related (9)
| Interface | Function |
|------|------|
| `product_detail` | Product Details |
| `product_variations` | Variation Details |
| `product_trend` | Sales/Price/Ranking Trend |
| `product_reviews` | User Reviews |
| `product_traffic_terms` | Traffic Keywords |
| `competitor_product_keywords` | Competitor Keyword Layout |
| `product_search` | Product Search Filter |
| `potential_product_search` | Potential Product Mining |
#### Category Related (7)
| Interface | Function |
|------|------|
| `category_report` | Category Real-time Report |
| `category_trend` | Category Trend Analysis |
| `category_keywords` | Category Core Keywords |
| `category_market_search` | Category Market Search |
#### Keyword Related (4)
| Interface | Function |
|------|------|
| `keyword_detail` | Keyword Details |
| `keyword_trend` | Keyword Trend |
| `keyword_related_words` | Long-tail Keyword Mining |
| `keyword_search_result` | Search Result Ranking |
---
## Project Structure
```
amazon-mcp/
├── .mcp.json # MCP Configuration File
├── .claude/
│ └── skills/
│ └── amazon-analyse/ # Competitive Product Analysis Skill
│ ├── SKILL.md # Skill Main Document
│ └── references/
│ └── sorftime-mcp-api.md # API Complete Documentation
├── reports/ # Analysis Report Directory
│ ├── analysis_XXX_US_YYYYMMDD.md
│ └── archive/
│ ├── 2025/
│ └── 2024/
└── README.md # This Document
```
---
## Analysis Report Instructions
### Report Structure
Each analysis report contains the following sections:
1. **Product Basic Data** - Price, Rating, Ranking, Sales
2. **Keyword Layout Analysis** - Traffic Keywords, Competitor Layout, Copywriting Strategy
3. **Review Qualitative Analysis** - Top 3 Advantages/Pain Points, Improvement Suggestions
4. **Competitive Strategy Analysis** - SWOT Analysis, Market Opportunities
5. **Strategic Counterattack Suggestions** - Keyword, Pricing, Product, Listing Optimization
### Report Naming Conventions
```
analysis_{ASIN}_{Site}_{Date}.md
Example: analysis_B07PWTJ4H1_US_20260302.md
```
### Report Management
| Stage | Time Range | Location |
|------|----------|------|
| Active Period | Last 30 Days | `reports/` |
| Reference Period | 1-6 Months | `reports/archive/YYYY/` |
| Archival Period | 6 Months+ | Can be deleted or compressed |
---
## Common Issues
### Q: Where is the API Key configured?
A: Edit the `.mcp.json` file and replace `YOUR_API_KEY` with your Sorftime API Key.
sorftime mcp key address:
https://sorftime.com/zh-cn/mcp
### Q: What analysis dimensions are supported?
A: Supports 15+ dimensions of analysis, including:
- Product Basic Information
- Sales/Price Trends
- User Review Sentiment
- Traffic Source Analysis
- Competitor Keyword Layout
- Category Market Analysis
- Long-tail Keyword Mining
### Q: How long does the analysis take?
A: Usually 30-60 seconds, depending on the amount of data and network conditions.
### Q: Can the report be exported?
A: The report is in Markdown format and can be converted to PDF, HTML, Excel.
---
## Skill Development
### Create New Skill
Use the `skill-creator` skill to quickly create:
```bash
# Initialize a new skill template
.claude/skills/skill-creator/scripts/init_skill.py <skill-name> --path .claude/skills
# Package the skill into a .skill file
.claude/skills/skill-creator/scripts/package_skill.py <skill-folder>
```
### Skill Design Principles
1. **YAML frontmatter** - Contains a complete `description` explaining the use case
2. **Progressive disclosure** - Keep SKILL.md concise, put details in references/
3. **Scripts** - For code that needs deterministic execution
4. **References** - Store API documentation, guides, and other reference materials
5. **Assets** - Templates, images, etc. required for output files
---
## Update Log
### v2.1 (2026-03-02)
- Added complete Sorftime MCP API documentation
- Added a comparison table of research dimensions and interfaces
- Optimized report structure
### v2.0 (2026-03-02)
- Redesigned the analysis framework
- Added four major dimension analysis models
- Improved report output format
### v1.0 (Initial Version)
- Basic competitive product analysis function
- Sorftime MCP integration
---
## License
MIT License
---
## Contact Information
- Sorftime Official Website: https://www.sorftime.com
- Claude Code Documentation: https://claude.ai/code
---
*Last Updated: 2026-03-02*
