功能描述

Amazon keyword research and market opportunity analysis for sellers. Retrieve autocomplete suggestions (long-tail keywords), analyze competitor landscape, an...

使用说明 (SKILL.md)

Amazon Keyword Research 🔍

Name: Amazon Keyword Research
Author: phheng

Free keyword research for Amazon sellers. No API key — works out of the box.

Installation

npx skills add nexscope-ai/Amazon-Skills --skill amazon-keyword-research -g

Capabilities

Long-tail keyword mining: Extract 100-200 real search terms from Amazon's autocomplete engine
Competitor landscape analysis: Product count, price range, average rating, review distribution, top brands
Seasonal trend detection: 12-month Google Trends data to identify peak seasons and demand shifts
Market opportunity scoring: 1-10 score combining competition density, price room, and demand signals
Multi-marketplace support: US, UK, DE, FR, IT, ES, JP, CA, AU, IN, MX, BR
Keyword comparison: Side-by-side analysis of multiple keywords

Usage Examples

Users can ask naturally. Examples:

Research the keyword "portable blender" on Amazon US

Find long-tail keywords for "yoga mat" on Amazon

I want to sell resistance bands. What does the Amazon keyword landscape look like?

Compare "laptop stand" vs "monitor stand" on Amazon US — which has more opportunity?

Analyze "Küchenmesser" on Amazon Germany

Research "water bottle" across Amazon US, UK, and DE

Workflow

Step 1: Gather Autocomplete Data

Run the bundled script to collect Amazon autocomplete suggestions:

\x3Cskill>/scripts/research.sh "\x3Ckeyword>" [marketplace]

Parameters:

keyword (required): The seed keyword to research
marketplace (optional): us (default), uk, de, fr, it, es, jp, ca, au, in, mx, br

What the script does:

Queries Amazon's autocomplete API with the seed keyword
Expands with prefixes: "best [keyword]", "cheap [keyword]", "top [keyword]"
Expands with a-z suffixes: "[keyword] a", "[keyword] b", ... "[keyword] z"
Returns deduplicated, sorted list of real search suggestions — one per line

Why this matters: Amazon autocomplete reflects what real shoppers are actually typing. These aren't guesses — they're demand signals directly from Amazon's search engine. The prefix and alphabet expansion catches long-tail terms that basic autocomplete misses, which are often lower competition and higher intent.

Example:

\x3Cskill>/scripts/research.sh "portable blender" us
# Returns 100-200 long-tail keywords

For multi-marketplace research, run the script once per marketplace.

Step 2: Analyze Competition

Use web_search to gather competitor intelligence:

Search "\x3Ckeyword>" site:amazon.com — note approximate result count for competition density
Search "\x3Ckeyword>" amazon best sellers price review — extract price patterns, rating averages, dominant brands
Summarize: total competitors, price range (min/avg/max), average star rating, top 5 brands by visibility

Why this matters: Raw keyword volume means nothing without competition context. A keyword with 10,000 searches but dominated by 3 entrenched brands with 10,000+ reviews each is a very different opportunity than one with the same volume but fragmented sellers. The price range reveals margin potential — if everything is under $10, margins will be razor-thin after FBA fees.

Step 3: Check Seasonality

Use web_fetch on Google Trends:

https://trends.google.com/trends/explore?q=\x3Ckeyword>&geo=US

If Google Trends returns a 429 error, fall back to web_search for seasonal data:

"\x3Ckeyword>" seasonal trends demand peak months

Identify: trend direction (rising/declining/stable), seasonal peaks (which months), year-over-year change.

Why this matters: Seasonality determines cash flow risk. A product that sells 80% of its volume in Q4 means you need capital for inventory months in advance and may sit on dead stock the rest of the year. Rising trends mean growing demand and more room for new entrants; declining trends mean you're fighting over a shrinking pie. This context turns a keyword from a number into a business decision.

Step 4: Synthesize Report

Combine all data into the output format below.

Why structure matters: Grouping keywords by intent (commercial vs informational vs niche) helps the seller understand not just what people search, but why they search it. The opportunity score condenses multiple signals into a single actionable number, but the breakdown behind it is what actually informs the decision — so always show the reasoning.

Output Format

Present the final report in this structure:

## Keyword Research Report: [keyword]
**Marketplace:** Amazon [US/UK/DE/...]
**Date:** [current date]

### 1. Long-tail Keywords ([count] found)

**High Commercial Intent:**
- [keyword with "buy", "best", "vs", "for" etc.]
- ...

**Informational / Research:**
- [keyword with "how to", "what is", "review" etc.]
- ...

**Niche / Specific:**
- [long, specific keywords indicating clear purchase intent]
- ...

### 2. Competition Landscape

| Metric | Value |
|--------|-------|
| Estimated competitors | [number] |
| Price range | $[min] - $[max] |
| Average price | $[avg] |
| Average rating | [stars] |
| Top brands | [brand1, brand2, brand3...] |

### 3. Seasonal Trends

[Describe 12-month trend: peaks, valleys, stable periods]
[Note any upcoming peak seasons relevant to the keyword]

### 4. Market Opportunity Score: [X/10]

**Score breakdown:**
- Competition density: [low/medium/high] — [why]
- Price room: [low/medium/high] — [why]
- Demand trend: [growing/stable/declining] — [why]
- Niche potential: [low/medium/high] — [why]

**Recommendation:** [1-2 sentence actionable recommendation]

Multi-Keyword Comparison

When the user asks to compare two or more keywords, run the full workflow (Steps 1-4) for each keyword separately, then present results in a side-by-side comparison table.

Example user input:

Compare "laptop stand" vs "monitor stand" vs "tablet stand" on Amazon US — which one should I sell?

How to execute: Run the script 3 times:

\x3Cskill>/scripts/research.sh "laptop stand" us
\x3Cskill>/scripts/research.sh "monitor stand" us
\x3Cskill>/scripts/research.sh "tablet stand" us

Then complete Steps 2-3 for each keyword, and output a comparison table:

Metric	laptop stand	monitor stand	tablet stand
Long-tail count	—	—	—
Avg price	—	—	—
Top brand dominance	—	—	—
Trend direction	—	—	—
Opportunity score	—	—	—

End with a Recommendation stating which keyword has the best opportunity and why.

Limitations

This skill uses publicly available data (Amazon autocomplete + web search). It does not provide exact monthly search volumes or sales estimates. For precise data, stay tuned for Nexscope — coming soon.

Part of the Nexscope suite — AI-powered Amazon seller tools.

安全使用建议

This skill appears to implement the advertised Amazon keyword research features, but there are a few red flags you should resolve before installing or trusting it: (1) The bundled script uses curl and python3 but the skill metadata does not declare these dependencies — ensure your environment has them and ask the author to declare them. (2) SKILL.md shows an npx install example that points to an external repo; confirm that source is legitimate before running the command. (3) The instructions tell the agent to trigger the skill on many vague user prompts — if you don't want frequent automatic scraping or network calls, limit or remove those trigger rules. (4) Because the skill performs web scraping and network requests (Amazon completion endpoints and Google Trends), review network access policies and be mindful of rate limits/ToS. If you need higher assurance, ask the publisher for: an explicit install spec, declared binary/runtime requirements, and a more narrowly scoped trigger policy.

功能分析

Type: OpenClaw Skill Name: amazon-keyword-research Version: 0.1.0 The skill provides legitimate Amazon keyword research functionality but contains a shell injection vulnerability in `scripts/research.sh`. The script passes the user-provided keyword directly into a `python3 -c` command string using single quotes without sanitization, which allows for arbitrary command execution if the input contains a single quote (e.g., `' ; touch /tmp/pwned ; '`). While the behavior aligns with the stated purpose and there is no evidence of intentional malice or data exfiltration, the high-risk nature of this vulnerability meets the criteria for a suspicious classification.

能力评估

ℹ Purpose & Capability

The script + SKILL.md support the advertised features (autocomplete scraping, alphabet expansion, web-based competitor and seasonality checks). However the skill fails to declare that it depends on curl and python3 (the bundled script calls both), and the README shows an npx install example pointing to an external repo even though the skill has no install spec — these mismatches are unexpected.

⚠ Instruction Scope

Runtime instructions tell the agent to run the included scraping script and to use web_search/web_fetch for competitor and Google Trends data, which is appropriate. But the SKILL.md also instructs the agent to 'make sure to use this skill whenever the user mentions' a long list of Amazon-related phrases (including vague questions). That is overbroad scope creep and may cause the agent to invoke the skill unexpectedly or on user queries where this level of web scraping isn't appropriate.

ℹ Install Mechanism

No formal install spec is present (lowest-risk pattern). The SKILL.md includes an npx command as an example that references a third-party package (nexscope-ai/Amazon-Skills). Because the registry metadata/source is unknown, that example is misleading and could send users to install external code; this should be clarified before following it.

⚠ Credentials

The skill requests no credentials or env vars, which is appropriate. However the included script requires system tools (curl and python3) that aren't declared in the skill metadata's 'required binaries' list. That discrepancy could lead to runtime failures or surprises. The skill does not request unrelated secrets, which is good.

ℹ Persistence & Privilege

The skill is not marked always:true and does not request persistent system modifications. However, the SKILL.md's insistence that the agent always trigger this skill for many user utterances increases the chance of autonomous invocation; treat that as a behavioral risk rather than a metadata privilege.

版本历史

v0.1.0

Initial release of Amazon Keyword Research skill: - Instantly generates 100-200 long-tail keyword suggestions using Amazon autocomplete, with expansions and deduplication. - Provides competitor landscape analysis: counts, price range, review distribution, and top brands. - Assesses seasonal demand using 12-month Google Trends data. - Calculates a market opportunity score (1–10) based on competition, price room, demand trend, and niche potential. - Supports 12 Amazon marketplaces (US, UK, DE, FR, IT, ES, JP, CA, AU, IN, MX, BR) without requiring an API key. - Includes structured, actionable reports and side-by-side keyword comparison.

元数据

Slug amazon-keyword-research

版本 0.1.0

许可证 MIT-0

累计安装 3

当前安装数 3

历史版本数 1

常见问题