← 返回 Skills 市场

Amz Product Optimizer

Name: Amz Product Optimizer
Author: simoncai519

作者 simoncai519 · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

122

总下载

当前安装

版本数

在 OpenClaw 中安装

/install amz-product-optimizer

功能描述

Use when user wants to optimize Amazon product listings, generate product images, improve titles, monitor CTR, or automate end-to-end Amazon product optimiza...

使用说明 (SKILL.md)

Amazon Product Optimizer (AMZ Product Optimizer)

Overview

Automate the complete Amazon product optimization pipeline: hot keyword extraction, title generation, detail image creation, and main‑image click‑through‑rate (CTR) monitoring. Suitable for sellers, product managers, and cross‑border e‑commerce teams.

Core Workflow

Get Hot Keywords – Scrape top keywords for the target niche.
Read Product Data – Load products from a local JSON/CSV file.
Optimize Titles – Generate fluent titles using the standard structure [Brand] + [Core Keyword] + [Feature] + [Spec].
Generate Detail Images – Create five scene‑based images via Taobao MCP.
Save Results – Update the product file with new titles and image URLs.
CTR Monitoring (optional) – Schedule daily checks of main‑image CTR, flagging products below the 5 % threshold.

Execution Modes

Mode	Description	Typical Use
`full`	Run the entire pipeline end‑to‑end.	First‑time bulk optimization
`keywords_only`	Retrieve hot keywords only.	Market research
`optimize_names`	Optimize product titles only.	When titles need refresh
`generate_images`	Produce detail images only.	After titles are final
`monitor`	Perform CTR monitoring only.	Daily health check

Parameters

keyword (string, required) – The niche or product category, e.g. "cat food".
product_file (string, required) – Path to a JSON/CSV containing product records.
mode (string, optional, default "full") – One of the modes above.

Usage Examples

User: Optimize my cat food listings
Assistant: (mode: full, keyword: "cat food", product_file: "products.json")

User: Show me hot keywords for dog beds
Assistant: (mode: keywords_only, keyword: "dog bed")

User: Generate detail images for product ID 12345
Assistant: (mode: generate_images, product_file: "products.json", keyword: "cat food")

User: Monitor main‑image CTR daily
Assistant: (mode: monitor)

Reference Files

references/optimization-guide.md – Detailed strategies for title optimization, image prompts, and CTR thresholds.
scripts/ – Helper scripts for keyword scraping, title generation, image creation, and monitoring (may be added later).

Notes

Titles must follow the standard structure and avoid keyword stuffing or "Hot Search" markers.
Image prompts are scene‑based to maximize conversion.
CTR threshold recommended at 5 %; adjust via configuration.
All results are saved back to the provided product_file.

Created by Simon Cai · More e-commerce skills: github.com/simoncai519/open-accio-skill

安全使用建议

Before installing or running this skill: (1) Ask the author to provide an explicit list of required credentials (Taobao/MCP, and Amazon/Seller Central or analytics) and how they will be used and stored. (2) Request a concrete install/runtime plan (how Python deps are installed, what scripts run, and where network calls go). (3) Do not hand over Amazon/Seller-Central credentials or broad API keys until you (or a reviewer) can inspect the code or scripts that will use them. (4) Clarify how CTR data is retrieved (API vs scraping) and whether the agent will perform scheduled background tasks — require explicit opt-in for daily monitoring. (5) Prefer running initial tests on a copy of your product file in an isolated environment. If the author cannot provide clear, verifiable details (code, install steps, or explicit credential requirements), treat the skill as untrusted.

功能分析

Type: OpenClaw Skill Name: amz-product-optimizer Version: 1.0.0 The skill bundle describes a legitimate e-commerce automation workflow for Amazon product optimization, including keyword scraping, title generation, and image creation via a Taobao MCP service (ID: 19cf03a191f). The requested access to local product files and external dependencies (requests, beautifulsoup4) is consistent with the stated purpose, and no malicious instructions, exfiltration logic, or obfuscation were found in the SKILL.md or documentation.

能力评估

⚠ Purpose & Capability

The skill claims to call Taobao MCP for image generation and to monitor main-image CTR (which would normally require Amazon/Seller-Central access), yet the package declares no required environment variables, credentials, or endpoints. The inclusion of a Taobao service ID in documentation without any declared credential handling is inconsistent with the stated capabilities.

⚠ Instruction Scope

SKILL.md instructs the agent to 'scrape top keywords', 'generate images via Taobao MCP', read/write local product JSON/CSV, and 'schedule daily checks' for CTR. It does not specify which endpoints/APIs to use for keyword scraping or CTR retrieval nor how credentials should be provided or protected, leaving broad discretion to the agent and unclear data flows.

ℹ Install Mechanism

This is instruction-only (no install spec or code files). The references list Python packages and an MCP service ID but there are no concrete install steps. While instruction-only reduces immediate disk-write risk, the missing install and dependency guidance is inconsistent with the declared Python dependencies and expected runtime actions.

⚠ Credentials

No environment variables or primary credentials are declared, yet operations described (Taobao MCP calls, CTR monitoring likely against Amazon analytics) normally require API keys/credentials. The lack of declared credentials is disproportionate to the external services the skill intends to use.

⚠ Persistence & Privilege

The skill suggests scheduling daily CTR monitoring and updating product files, implying ongoing background activity. Although it does not set always:true, autonomous invocation plus the unclear mechanism for scheduling and data access increases the risk of persistent network activity or repeated file writes without explicit user consent.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install amz-product-optimizer
安装完成后，直接呼叫该 Skill 的名称或使用 /amz-product-optimizer 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

- Added a version field to explicitly specify version "1.0.0" in the metadata. - Moved repository attribution to the end and reformatted it for clarity. - No changes to functionality or workflow; documentation revision and metadata update only.

v0.1.1

- Added a repository attribution notice at the top of the SKILL.md file, referencing the source and providing a link for more e-commerce skills. - No changes to functionality or features; only documentation updated.

v0.1.0

initial release

元数据

Slug amz-product-optimizer

版本 1.0.0

许可证 MIT-0

累计安装 1

当前安装数 1

历史版本数 3

常见问题