Install
openclaw skills install @jinhuadeng/xianyu-auto-opsBilingual Xianyu (闲鱼) listing and lightweight operations workflow for second-hand goods, side-hustle products, and marketplace distribution. Use when the user wants to create, optimize, or batch-produce Xianyu listing assets such as titles, selling points, product descriptions, image prompts, reply scripts, pricing angles, posting checklists, or simple operating SOPs in Chinese and English. Also use when the user asks to turn product info or CSV-like SKU data into publish-ready marketplace materials, wants category-specific Xianyu templates, needs buyer chat replies, or wants a reusable batch-oriented Xianyu sales process for physical goods, digital products, or AI services.
openclaw skills install @jinhuadeng/xianyu-auto-opsUse this skill to turn rough product information into a repeatable Xianyu / Idle Fish operating package.
Default output language: bilingual Chinese + English.
Default business goal: faster listing, clearer positioning, better inquiry conversion.
Follow this sequence unless the user asks for only one part.
Clarify the offer
Choose the operating mode
Pick a category template Use the closest category framing:
references/ai-services-template.md when the product is training, installation, deployment, consulting, or AI service resale.Produce the listing package Return, in this order when relevant:
Keep it platform-native
When the user gives multiple items, default to a compact table-like structure using bullets, not markdown tables on chat surfaces.
For each SKU, include:
If the user provides a CSV or spreadsheet, use scripts/batch_csv_to_brief.py <file> to normalize the rows first, then use the JSON output as batch input.
Expected columns can include:
sku, name, category, brand, condition, price_target, flaws, accessories, city, delivery, notes
Chinese headers like 商品名 / 类目 / 成色 / 价格 / 瑕疵 / 配件 / 城市 / 发货 / 备注 are also supported.
Do not ask the user to perfect the data before starting. Fill gaps with assumptions and mark them.
Write titles that are:
Prefer this rough formula:
[brand/category] + [core item] + [condition / key value] + [buyer use case / bonus point]
For each Chinese description, keep this structure:
For English, provide a shorter mirror summary rather than a full literal translation unless the user asks for full bilingual parity.
When suggesting price, provide three layers when possible:
Base suggestions on:
When generating reply scripts, include short ready-to-send messages for:
Keep replies short and human.
When the user asks for ad visuals, output two prompt layers:
Prefer 16:9 for article covers and 1:1 / 4:5 for feed-like visuals unless the user says otherwise.
Use this template unless the user asks for a different one.
中文版本
English version
When the user asks for bilingual output, do not translate mechanically.
Read references/playbook.md when the user needs stronger title formulas, batch handling, category-specific patterns, buyer reply banks, or poster prompt templates.