基于证据改善商品描述的信息传达
Coding
Amazon 商品 FAQ 建议
Try it从评论疑问提炼商品页 FAQ 建议
What it does
从 Amazon 评论中的重复疑问和商品字段中提炼商品页 FAQ 主题与回答证据。仅用于 FAQ 主题建议;不用于客服自动回复、法律审核或自动发布。 Requires an ARI API key (ari_live_*).
The skill document
FAQ Generator
Creates FAQ lists from medical documents.
Features
- Automatic Q&A generation
- Policy interpretation
- Patient-friendly language
- Structured formatting
Parameters
| Parameter | Type | Default | Required | Description |
|---|---|---|---|---|
--input, -i | string | - | Yes | Source document file path |
--audience, -a | string | general | No | Target audience (patients, researchers, general) |
--output, -o | string | stdout | No | Output file path |
--format, -f | string | json | No | Output format (json, markdown, text) |
Output Format
{
"faqs": [{"question": "", "answer": ""}],
"topic": "string"
}
Risk Assessment
| Risk Indicator | Assessment | Level |
|---|---|---|
| Code Execution | No scripts included | Low |
| Network Access | No external API calls | Low |
| File System Access | Read-only within workspace | Low |
| Instruction Tampering | Standard prompt guidelines | Low |
| Data Exposure | Input/output within session | Low |
Security Checklist
- No hardcoded credentials or API keys
- No unauthorized file system access (../)
- No network requests to external services
- Output does not expose sensitive information
- Prompt injection protections in place
Evaluation Criteria
Success Metrics
- FAQ accurately represents source document content
- Language is appropriate for specified audience (patients/researchers)
- Questions cover key points of the document
- Answers are clear, concise, and medically accurate
- Format follows structured JSON schema
Test Cases
- Basic FAQ Generation: Input simple medical protocol → Output valid FAQ list
- Audience Adaptation: Same input with different audiences → Appropriate tone shift
- Complex Document: Input lengthy policy document → Comprehensive FAQ coverage
- Edge Case: Input ambiguous content → Handles gracefully with clarifying questions
Lifecycle Status
- Current Stage: Draft
- Next Review Date: 2026-03-06
- Known Issues: None
- Planned Improvements:
- Add support for multi-language output
- Enhance medical terminology handling
Related skills
将评论痛点整理为可验证产品需求
从差评与低星体验定位质量改进线索
把评论问题转成产品、包装与供应商任务
从评论识别可能导致退货的体验线索
找出页面表达与真实买家体验的差距