agentleFS
Ethan-paseyaEthan-paseya/amazon-sellersprite-research-MCP-skill/.claude/skills/product-research/SKILL.md

product-research

Skill2 starsChanged 4 months agoMITView on GitHub

Amazon product opportunity research using SellerSprite MCP and LLM reasoning. Use when the user runs `/product-research "{PRODUCT_KEYWORD}" {SITE}` to research a product direction, evaluate market opportunity, competitors, VOC, barriers, and launch decision.

14 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 13 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Before any data call, read references/runtime-credential-preflight.md when it is available.

    line 20
  2. Distinctive:

    If SellerSprite MCP is not configured, unavailable, or returns an authentication error, ask the user for the SellerSprite MCP API key …

    line 22
  3. Distinctive:

    If Gemini/GLM validation is part of the requested output and model keys are missing, ask for the missing keys before validation.

    line 23
  4. Distinctive:

    Do not continue with mock MCP data or fabricated opportunity conclusions.

    line 24
  5. Distinctive:

    Do not write provided keys into reports, raw files, dashboards, examples, or the open-source repository.

    line 25
  6. Very specific:

    Read references/sellersprite-mcp-api.md before execution when it is available.

    line 29
  7. Distinctive:

    If MCP cannot answer a required question, mark the gap and then use a crawl/browser fallback only for that missing evidence.

    line 46
  8. Distinctive:

    Do not invent supply-chain, patent, compatibility, seasonality, inventory, or certification risks.

    line 71
  9. Distinctive:

    Do not publish raw product exports or private cost assumptions.

    line 81

“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Product Research

Command

/product-research "{PRODUCT_KEYWORD}" {SITE}

Goal

Evaluate a product opportunity and produce a decision-ready research report.

Credential Preflight

Before any data call, read references/runtime-credential-preflight.md when it is available.

  • If SellerSprite MCP is not configured, unavailable, or returns an authentication error, ask the user for the SellerSprite MCP API key before product research.
  • If Gemini/GLM validation is part of the requested output and model keys are missing, ask for the missing keys before validation.
  • Do not continue with mock MCP data or fabricated opportunity conclusions.
  • Do not write provided keys into reports, raw files, dashboards, examples, or the open-source repository.

Required Reference

Read references/sellersprite-mcp-api.md before execution when it is available. Normalize GB to SellerSprite MCP UK before tool calls.

Data Calls

  • product_node: map the product direction to relevant Amazon category nodes.
  • product_research: collect candidate products and opportunity-filtered samples.
  • competitor_lookup: compare representative products by keyword, ASIN, brand, seller, node, month, and variation setting.
  • market_research and market_research_statistics: quantify category market size and competition.
  • market_price_distribution, market_rating_distribution, market_product_demand_trend: check pricing, maturity, and demand trend.
  • market_product_concentration, market_brand_concentration, market_seller_concentration: quantify head concentration.
  • keyword_miner, keyword_research, keyword_research_trends: validate keyword demand, CPC/PPC, purchase rate, supply-demand ratio, and trend.
  • traffic_listing, traffic_source, traffic_extend: inspect traffic patterns for representative competitors when available.
  • review: collect VOC from representative ASINs.
  • google_trend: optional external demand validation.
  • trademark_country_list, trademark_list, trademark_stats, trademark_detail: use only for naming/IP/trademark risk checks.

If MCP cannot answer a required question, mark the gap and then use a crawl/browser fallback only for that missing evidence.

Workflow

  1. Clarify product keyword and site.
  2. Normalize site code for SellerSprite MCP.
  3. Collect representative products using product_research and competitor_lookup.
  4. Collect keyword, category, market distribution, and traffic context from MCP.
  5. Save raw or compact MCP responses before reasoning.
  6. Label product attributes: price, rating, reviews, feature, material, scenario.
  7. Compare opportunity patterns with evidence references.
  8. Analyze VOC and barriers using review/category/trademark evidence or labeled assumptions.
  9. Output decision: enter, observe, or reject.

Research Dimensions

DimensionQuestions
DemandIs there enough search and sales demand?
CompetitionAre review moats and brands too strong?
DifferentiationCan the product be meaningfully improved?
ProfitIs price/CPC/return risk acceptable?
Supply chainIs production feasible?
Launch pathWhat keywords and segments should be tested first?

Do not invent supply-chain, patent, compatibility, seasonality, inventory, or certification risks. Use MCP data first, then existing user files, then explicitly labeled assumptions or crawl evidence.

Output

Save under:

product-research-reports/{PRODUCT_KEYWORD}_{SITE}_{YYYYMMDD}/report.md

Do not publish raw product exports or private cost assumptions.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.