APIClaw — Amazon Seller Data Analysis

AI-powered Amazon product research. Respond in user's language.

Files

Credential

Required: APICLAW_API_KEY. Get free key at apiclaw.io/api-keys. Stored in {skill_base_dir}/config.json in skill root.

Input

User provides: keyword, category, ASIN, or brand — depending on intent. Use intent routing below.

API Pitfalls (CRITICAL)

1. Category first: keyword search is broad → MUST lock categoryPath via categories endpoint before other calls
2. Brand + category: Brand queries MUST include --category to avoid cross-category contamination
3. Use API fields directly: revenue=sampleAvgMonthlyRevenue (NEVER calculate price×sales), sales=monthlySalesFloor (lower bound), opportunity=sampleOpportunityIndex
4. reviews/analysis: needs 50+ reviews per ASIN; try category mode first (single call returns all dimensions), ASIN mode only if category call fails. Filter by labelType client-side from the consumerInsights array.
5. Aggregation without categoryPath: produces severely distorted data
6. .data is array: use .data[0], not .data.field
7. labelType: NOT an API request parameter — it is a field in the response consumerInsights array, used for client-side filtering
8. history empty: try oldest-listed ASINs first, up to 3 rounds of different ASINs before giving up
9. Sales null fallback: Monthly sales ≈ 300,000 / BSR^0.65

14 Product Selection Modes

Modes can combine with explicit filters (--price-max, --sales-min, etc). Overrides win.

Composite Commands

• report --keyword X → categories + market + products(top50) + realtime(top1)
• opportunity --keyword X [--mode Y] → categories + market + products(filtered) + realtime(top3)

Analysis Framework

Every analysis should address these dimensions where data is available:

Market Health Assessment

Competitive Position Assessment

• Price vs category avg: >20% above = premium positioning, >20% below = value play 🔍
• Rating vs category avg: ≥0.3 above = quality advantage, ≥0.3 below = quality risk 🔍
• Review count vs Top 10 avg: \x3C10% of leaders = high barrier, >50% = competitive 🔍
• BSR trend (30d): Improving = momentum, stable = holding, declining = losing share 🔍

Opportunity Viability

When user asks "should I sell X" or "is this a good niche":

• ALL of: demand >500, CR10 \x3C60%, avgReviewCount \x3C5,000 → Likely viable 🔍
• ANY of: demand \x3C200, CR10 >80%, avgReviewCount >10,000 → Likely not viable 🔍
• Mixed signals → Present data, let user decide with their domain knowledge 💡

Sales Estimation Notes

• monthlySalesFloor is a lower-bound estimate 📊
• Null sales fallback: Monthly sales ≈ 300,000 / BSR^0.65 🔍
• Revenue = sampleAvgMonthlyRevenue directly — NEVER calculate price × sales 📊

Output Spec

Sections: Analysis findings → Data Source & Conditions table (interfaces, category, dateRange, sampleType, topN, filters) → Data Notes (estimated values, T+1 delay, sampling basis).

Language (required)

Output language MUST match the user's input language. If the user asks in Chinese, the entire report is in Chinese. If in English, output in English. Exception: API field names (e.g. monthlySalesFloor, categoryPath), endpoint names, technical terms (e.g. ASIN, BSR, CR10, FBA, credits) remain in English.

Disclaimer (required, at the top of every report)

Data is based on APIClaw API sampling as of [date]. Monthly sales (monthlySalesFloor) are lower-bound estimates. This analysis is for reference only and should not be the sole basis for business decisions. Validate with additional sources before acting.

Confidence Labels (required, tag EVERY conclusion)

• 📊 Data-backed — direct API data (e.g. "CR10 = 54.8% 📊")
• 🔍 Inferred — logical reasoning from data (e.g. "brand concentration is moderate 🔍")
• 💡 Directional — suggestions, predictions, strategy (e.g. "consider entering $10-15 band 💡")

Rules: Strategy recommendations are NEVER 📊. Anomalies (>200% growth) are always 💡. User criteria override AI judgment.

Data Provenance (required)

Include a table at the end of every report:

Extract endpoint and params from _query in JSON output. Add notes: sampling method, T+1 delay, realtime vs DB, minimum review threshold, etc.

API Usage (required)

Extract from meta.creditsConsumed per response. End with Credits remaining: N.

Limitations

Cannot do: keyword research, reverse ASIN, ABA data, traffic source analysis, historical price/BSR charts. Niche keywords may return empty — use category path instead.