launchfast-product-research
|
它会碰到什么
扫了多少1 个文本文件,2 KB
它会碰到什么不碰外部(只输出文字)
命中总数0 处
命中统计严重 0 · 高 0 · 中 0 · 低 0
这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。
技能内容
LaunchFast Product Research
Use this skill for a fast, opinionated market scan across up to 10 keywords.
Inputs
If missing, ask once for:
- keywords to research
- optional target price band
- optional competition tolerance
Workflow
1. Run research
- Run
research_productsfor each keyword. - Prefer parallel tool calls when researching more than one keyword.
- Use
focus="balanced"andproduct_limit=20unless the user asks otherwise.
2. Extract core metrics per keyword
From the response, compute:
- products analyzed
- grade distribution
- median monthly revenue
- top product revenue
- average or median price
- average or median reviews
- brand concentration
- dominant brand
3. Score the market
Use this score out of 100:
score =
(% of products graded B5 or better) * 30
+ (median revenue >= 8000 ? 30 : (median revenue / 8000) * 30)
+ (median reviews < 300 ? 20 : (300 / median reviews) * 20)
+ (median price between 18 and 60 ? 20 : 10)
Competition bands:
- Low: median reviews < 200
- Medium: median reviews 200-800
- High: median reviews > 800
Verdicts:
- GO: score >= 65
- INVESTIGATE: score 40-64
- PASS: score < 40
4. Present results
Always show:
- a ranked summary table for all keywords
- a short deep dive for the top 3
Use this table shape:
## Product Opportunity Scan
| Rank | Keyword | Score | Market Grade | Top Revenue | Avg Price | Competition | Verdict |
|------|---------|-------|--------------|-------------|-----------|-------------|---------|
For each top-3 keyword include:
- products analyzed
- grade distribution
- revenue range
- price range
- review range
- best product snapshot
- one-sentence key insight
- risk flags
- verdict with 1-2 sentence rationale
Notes
- If a keyword has missing or zero search volume, explicitly call that out.
- If the results look too narrow or too broad, suggest the better keyword variant inline.
- Do not force a follow-up question at the end; only recommend next actions when the result is borderline or strong enough to justify it.
想直接用这个技能?
本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。
它属于哪个仓库
星标★ 1,027
本站分层T1
该仓技能数1910
原文件路径
plugins/BlockchainHB/launchfast_codex_plugin/skills/launchfast-product-research/SKILL.md