SellerSprite Product Research

SellerSprite Product Research Skill — Input a keyword or category, automatically fetch market data via SellerSprite API and run AI analysis, output a structu...

Input a keyword or ASIN, AI automatically calls the SellerSprite API to fetch market data and outputs a structured product research report.

Prerequisites

Set your SellerSprite API Key (get one at open.sellersprite.com):

export SELLERSPRITE_SECRET_KEY="your-secret-key"

Ensure openclaw CLI is installed (used for AI analysis):

which openclaw

Quick Start

# Basic product research (keyword)
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "wireless earbuds"

# Specify marketplace (default: US)
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "yoga mat" --marketplace UK

# Use a specific AI model
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "phone case" --model claude
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "phone case" --model gemini
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "phone case" --model gpt-5
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "phone case" --model grok

# Use full model ID
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "yoga mat" --model anthropic/claude-opus-4-6

# Analyze a competitor by ASIN
bash ~/.claude/skills/sellersprite-product-research/selection.sh --asin B08N5WRWNW --marketplace US

# Save report to file
bash ~/.claude/skills/sellersprite-product-research/selection.sh --keyword "LED strip" --output report.md

Supported Models

Uses openclaw for AI inference. Pass --model with a shorthand alias or full model ID:

Alias	Model
claude	anthropic/claude-sonnet-4-6
claude-opus	anthropic/claude-opus-4-6
claude-haiku	anthropic/claude-haiku-4-5
gemini	google/gemini-2.5-pro
gemini-flash	google/gemini-2.5-flash
gpt-4o	openai/gpt-4o
gpt-5	openai/gpt-5
grok	xai/grok-4
deepseek	groq/deepseek-r1-distill-llama-70b
qwen	groq/qwen/qwen3-32b
mistral	mistral/mistral-large-latest

Any full model ID from openclaw models list --all is also accepted.

Sample Output

╔══════════════════════════════════════════════════════════════╗
║       SellerSprite Product Research Report                  ║
║  Keyword: wireless earbuds  |  Market: US  |  2026-03-09   ║
╚══════════════════════════════════════════════════════════════╝

📊 Market Overview
──────────────────────────────────────
  Products          1,284
  Avg Monthly Units 456/mo
  Avg Price         $28.50
  Avg Ratings       3,210
  Blue Ocean Index  ████░░░░░░░░░  3.2 / 10

🔴 Risk Signals
══════════════════════════════════════════════
1. High Competition — TOP10 monthly sales concentration 78%
2. Brand Barrier — Anker/JBL hold 35% market share
...

🟢 Opportunity Windows
══════════════════════════════════════════════
1. [Price Gap] $15-$20 range has few competitors but strong search volume
2. [New Product Bonus] New listings grew 42% in last 90 days
...

🎯 Recommended Entry Strategy
══════════════════════════════════════════════
1. Target $15-18 price range, differentiate on sport/waterproof features...

📌 Top Reference Products
──────────────────────────────────────────────
B0XXXXXXXX   2,340/mo   $19.9   4.3★   FBA  ⭐ Best Seller
...

How It Works

① Input keyword / ASIN / category
      ↓
② Call SellerSprite API (product research + keyword research)
      ↓
③ Parse market data (sales, price, competition, trends)
      ↓
④ Claude AI deep analysis (Blue Ocean scoring + strategy)
      ↓
⑤ Output structured product research report

Scripts

File	Description
selection.sh	Main entry point
fetch.sh	SellerSprite API data fetching
analyze.sh	AI analysis and report rendering

Marketplace Codes

Code	Market
US	United States
UK	United Kingdom
DE	Germany
JP	Japan
CA	Canada
FR	France
IT	Italy
ES	Spain
MX	Mexico
AU	Australia

Notes

• API rate limit: 40 requests/min, max 100 items per request
• Max 2,000 products per query
• Each analysis uses approx 3,000–8,000 tokens ($0.02–$0.05)
• Recommend querying data from the last 1–3 months for freshness