电商产品描述批量生成器

批量生成多平台电商产品标题、卖点和详情描述，支持中文英文及CSV导入，提升跨境及本土化文案效率。

Security Scan

The skill is internally coherent for a local, template-based product description generator: it asks for no credentials, has no network/install steps, and its CLI/tests match the documented functionality; the only concerns are minor code-quality items and that I could not fully review a truncated portion of generator.py.

Purpose & Capability

Name, description, README, SKILL.md, CLI and generator code are consistent: a local Python CLI produces platform-specific product descriptions and supports CSV batch input. No unrelated environment variables, binaries, or cloud credentials are requested.

Instruction Scope

Runtime instructions in SKILL.md and README are limited to running the CLI, reading a CSV for batch mode, and writing output files. There are no instructions to read arbitrary system files, access credentials, or send data to external endpoints.

Install Mechanism

No install spec is declared (instruction-only skill), and the repository uses pure Python with an optional setup.sh and a minimal requirements.txt (pytest). No downloads from third-party URLs or extracted archives are involved.

Credentials

The skill requires no environment variables or credentials. All inputs are CLI args or CSV files supplied by the user, which is proportionate to the stated purpose.

Persistence & Privilege

Flags show always:false (no forced inclusion). disable-model-invocation is false (normal). The code does not request system-wide configuration changes or other skills' credentials.

Assessment

This package appears to be a local, template-driven product-description generator and is coherent with its documentation: it doesn't ask for secrets or reach out to external services. Before installing or running it, consider: 1) Inspect generator.py fully (search for imports like requests, urllib, socket, subprocess, os.system, or any hardcoded endpoints) to confirm there are no hidden network calls. 2) Note setup.sh runs the test suite (pytest) which will execute code from the repository — run it in a sandbox/container if you want to limit risk. 3) The SKILL.md/README mention paid tiers and 'API' in marketing text but the code appears fully local; if you expect remote or subscription features, verify where those would be implemented. 4) If you plan to let an autonomous agent invoke this skill, be mindful that it can read any CSV you point it to and write files where it has filesystem permissions — avoid exposing sensitive files as inputs. If you want, I can grep the full generator.py for network/system-call patterns or review the rest of the file if you provide the remaining content.

ecommerce-product-desc-generator — 电商产品描述批量生成器

痛点

• 每次上新都要手动写标题、bullet points、详情描述，效率低下
• 同一产品在不同平台要写不同风格的文案，工作量翻倍
• 跨境卖家需要英文文案，中文卖家需要本土化表达
• 批量上新时，文案写作成为瓶颈

场景

• 亚马逊卖家：快速生成 SEO 友好的英文标题 + 5点描述 + 详情
• 淘宝/拼多多卖家：一键生成中文卖点的标题和详情页文案
• TikTok Shop 卖家：生成短视频风格的种草文案
• Shopify 独立站卖家：生成品牌感强的产品页面描述
• 批量上新：CSV 导入产品列表，一次生成全平台描述

定价

• Free（免费）：单产品单平台生成，基础模板
• Pro ¥29/月：批量生成（50产品/次），全5大平台，支持 CSV 导入导出
• Team ¥99/月：无限生成，API 调用，白标定制

指令格式

单产品生成

ecommerce-product-desc "产品名称" "品类" --keywords "关键词" --platforms amazon,tiktok --format markdown

全平台生成

ecommerce-product-desc "蓝牙耳机" "3C数码" --keywords "无线,降噪" --all-platforms

指定平台（逗号分隔）

--platforms amazon,taobao,pinduoduo,tiktok,shopify

输出格式

• --format markdown（默认）：每个平台一个 Markdown 区块
• --format txt：纯文本，平台间用 === 分隔
• --format csv：CSV 表格，含平台/标题/卖点/详情

批量生成（CSV）

ecommerce-product-desc --csv products.csv --format markdown --output result.md

CSV 格式（支持中文列名或英文列名）：

product_name,category,keywords,brand,price
蓝牙耳机,3C数码,无线,品牌X,199
充电宝,电子产品,大容量,品牌Y,89

批量模式（命令行）

ecommerce-product-desc --batch --product "充电宝" "电子产品" --all-platforms

支持平台

平台	语言	标题长度	特点
amazon	英文	≤200字符	5条Bullet Points + 详情段落，SEO优化
taobao	中文	≤30字	Emoji+卖点，淘宝风格
pinduoduo	中文	≤30字	低价引流词前置，拼团风格
tiktok	英文	≤100字	口语化+emoji，种草短视频风格
shopify	英文	品牌感	章节化描述，适合独立站

参数说明

参数	必填	说明
product	是	产品名称
category	否	品类/类目
--keywords	否	关键词，多个用逗号分隔
--brand	否	品牌名
--price	否	目标价格
--platforms	否	目标平台，默认 all
--format	否	markdown/txt/csv，默认 markdown
--output	否	输出文件路径
--seed	否	随机种子，用于复现结果