These six images are output sketches laid out like the real deliverables, not screenshots; swapping in real images keeps the same frame size.
Requires
What you need before running this pack. Miss one and it cannot do the work.
- Python python3 runs on your machine — one command, and it does not touch your files
- Report / exported CSV You export a CSV or JSON from the platform admin as the input
Picked most often by
Who buys it most, by role. Click a role to see only that role's skills.
Similar skills
View all →竞品拆解包
下架急救包(变狗 / 搜不到 / 受限商品)
欧盟合规包:上架前的四关自查(GPSR / PPWR / EPR / 欧代)
侵权自查与预警包
What is included
- ✓ reason-mix — reason-mix:退货原因 → 根因
- ✓ cost-model — cost-model:单笔退货成本
- ✓ listing-diff — listing-diff:把退货根因变成字段级改动
Who it is for
- ✓ People who need steady output in a fixed scene (Customer support / Finance & cost)
- ✓ You already have assets — you need structure
- ✓ Want to set it up once and reuse it, instead of re-explaining every time
Not for
- – Want full automation with no input at all
- – One-off use, no reuse planned
- – Platform-specific ad console actions (do those manually)
Use “退货降本包” to make me Web pages — I will paste the input below.
Copy this line to your agent to startAbout this skill
把一份退货报表跑成三样东西:按「我能改什么」分的六类根因(尺码/图文不符/破损/预期偏差/物流/其他)、每一单退货的 8 项成本与可手算的加法公式、以及改哪一段 listing 的字段级 diff(含尺码表与包装说明整文件)。三个 skill 全部离线、纯标准库、确定性输出,不联网也不需要平台授权。
Sample output
Real input → real output, unpolished# listing 字段级改动清单:GAN-TS-01(B0DEMO0001) 依据 **4 类根因**(尺码不符、图文不符、破损、预期偏差),覆盖该 SKU 退货的 **100%**;共 **10 段**要改。每段的「依据」都指向 `reason-mix-by-cause.csv` 里的原始原因值,可回查。 > 这不是策略,是 diff:下面每一段都写着「现在是什么 / 改成什么」,照着粘回后台即可。文案是模板 + 你的实测值确定性生成的,没有一句是模型编的。 ## 一页总表 | # | 优先级 | 字段 | 依据根因 | 占比 | 现在 | 改成 | |---|---|---|---|---|---|---| | 1 | P1 | 尺码表 / 尺寸表 | 尺码不符 | 73% | 未上传:Amazon 后台多数服装类目没有独立尺码表字段,尺码信息目前只写在一张图… | 见 after/size-chart.md(整张表重新生成,按实测值渲染) | | 2 | P1 | 五点第 3 条 | 尺码不符 | 73% | 宽松版型,不挑身材,遮肉显瘦 | 选码指引:S–XXL,胸围 96–120cm,平铺单面测量(衣服放平,从腋下 2c… | | 3 | P1 | 标题 | 尺码不符 | 73% | 棉质圆领短袖 T 恤 男女同款 柔软亲肤 四季可穿 送礼佳品 | 棉质圆领短袖 T 恤 男女同款 柔软亲肤 四季可穿 送礼佳品(S–XXL,胸围 9… | | 4 | P1 | 五点第 2 条 | 图文不符 | 9% | 柔软亲肤,四季可穿,送礼佳品 | 所见即所得:精梳棉 100%,180g/㎡,圆领短袖,6 个尺码(S–XXL)、5… | | 5 | P1 | 图片 1 的 alt | 图文不符 | 9% | 短袖正面平铺图 | 实物参照实拍:精梳棉 100%,180g/㎡,圆领短袖,6 个尺码(S–XXL)、… | | 6 | P1 | 描述(追加一段) | 图文不符 | 9% | 这款纯棉短袖是我们的经典款,面料柔软,做工扎实,适合日常穿着。多色可选,男女同款。
return_id,sku,product,units,cause,cause_zh,disposition,disposition_class,unit_price,unit_landed_cost,return_shipping,outbound_shipping,platform_fee,inspection,markdown,scrap,storage,cs_labor,total,share_of_price,formula,flag r001,GAN-TS-01,棉质圆领短袖 T 恤(男女同款),1,size,尺码不符,SELLABLE,sellable,39.00,14.80,5.90,3.85,2.00,1.45,10.92,0.00,0.63,1.98,26.73,0.6854,5.90+3.85+2.00+1.45+10.92+0.00+0.63+1.98=26.73, r002,GAN-HD-01,拉链连帽卫衣 抓绒内里,1,size,尺码不符,SELLABLE,sellable,62.00,24.50,5.90,3.85,2.00,1.45,17.36,0.00,0.63,1.98,33.17,0.5350,5.90+3.85+2.00+1.45+17.36+0.00+0.63+1.98=33.17, r003,GAN-TS-02,速干训练短袖 男款,1,size,尺码不符,SELLABLE,sellable,45.00,17.20,5.90,3.85,2.00,1.45,12.60,0.00,0.63,1.98,28.41,0.6313,5.90+3.85+2.00+1.45+12.60+0.00+0.63+1.98=28.41, r004,GAN-TS-01,棉质圆领短袖 T 恤(男女同款),1,size,尺码不符,SELLABLE,sellable,39.00,14.80,5.90,3.85,2.00,1.45,10.92,0.00,0.63,1.98,26.73,0.6854,5.90+3.85+2.00+1.45+10.92+0.00+
Install guide
One column per agent, each with a check that proves it installed. If the check fails, it is not installed.Claude Code
.claude/skills/returns-cost/
mkdir -p .claude/skills
unzip returns-cost.zip -d .claude/skills/
How to verify Open a new session and ask it to list the installed skills — seeing this pack's name means it worked.
For a global install, swap .claude/skills/ for ~/.claude/skills/ so every project of this user sees it.
Codex
~/.codex/skills/returns-cost/
mkdir -p ~/.codex/skills
unzip returns-cost.zip -d ~/.codex/skills/
How to verify Open a new codex session and ask it to list available skills — this pack showing up means it is live.
Codex only reads ~/.codex/skills/, not the project folder; on another machine you place it again.
Cursor
.cursor/rules/returns-cost.mdc
mkdir -p .cursor/rules
cp returns-cost/SKILL.md .cursor/rules/returns-cost.mdc
How to verify In chat, @ that rule file and ask what the pack's first hard rule is — a correct answer means it loaded.
Cursor has no separate skills folder: rules belong to the project, so you place them again per project — the biggest difference from the other two.
If it will not installNot opening a terminal at all? Paste the whole SKILL.md into the chat. Same result, you just re-paste it each new session.
Full steps and common errors live in the pack's playbook.md; the guides section also has ordering and usage guides。
How to use
4 steps; the first run takes about 10 minutes-
1
开工前准备 | 要做 | 怎么做 | 为什么 | |---|---|---| | 定一个时间窗口 | 最近 30 天或 90 天,退货表与销量表**用同一个窗口** | 分母不同期,退货率就是错的;大促后单独看,别混进日常 | | 导三份表 | ① 退货报表 ② 同期的已订购商品销售额(或店铺订单数) ③ 你的商品成本表(售价 + 落地成本) | 没有 ② 就算不出退货率;没有 ③ 第二关直接拒跑 | | 只拿一个类目先试 | 按类目筛,先 20–50 单退货 | 全店几千条一起跑,改稿清单会看不完 | | 确认退货原因列 | 用 Excel 打开,看有没有"原因"这一列,值长什么样(代码还是中文) | 归类全靠它;列名/取值对不上先去规则文件里加同义词 | | 备份 | 原始导出的 CSV 单独存一份 | 后面要重跑、要
-
2
装配 **Claude Code**:复制到项目的 `.claude/skills/`(没有就先建)。 **Codex CLI**:复制到 `.codex/skills/`。 **没有 shell**:不用装。把三个 `SKILL.md` 依次贴进对话,按里面写的规则手工执行;脚本那几步退化成口头执行,产出要你自己填进 CSV。**注意**:手工执行时第 3 关最容易走样(模板里的变量、路径要自己一个个对),建议还是装脚本。 **只用脚本不装 skill**:也可以。`skills/<名字>/scripts/*.py` 是独立的,`python3 <脚本> --help` 能看全部参数。 装完确认(一条动作):新开一个对话,问 **"列出你现在可用的 skills"**,应看到 `reason-mix`、`cost-
-
3
触发顺序 四步(第 1 步要跑两次)必须按顺序,原因不是习惯: 1. **先 `reason-mix` 跑全店,再按 SKU 跑一次**。因为第三关改的是**某一个 listing**,而全店的原因占比是多个 SKU 混在一起的。反着做会怎样:拿全店的"破损 46%"去改一个破损只占 9% 的服装 SKU,你会给它加气柱袋和易碎标——白花钱,真问题(尺码)一动不动。 2. **先 `reason-mix`,再 `cost-model`**。第二关的分组与"可售/报废"判定都来自第一关的产出。反着做会怎样:没有 `disposition_class`,报废与折价无法区分,成本会被系统性低估(样例 02 里 9 单有 7 单是整件报废,全按折价算会少算一半多)。 3. **先 `cost-model`,再 `listing-
-
4
每一步的输入与输出 ### 第 1 关 `reason-mix`(跑两次) | 项 | 内容 | |---|---| | 输入 | 退货报表 CSV(Amazon:`sku,quantity,reason,customer-comments,detailed-disposition`;Shopify:`sku,quantity,return_reason,return_note,warehouse_grade`)+ 销量表(可选) | | 命令 | `python3 skills/reason-mix/scripts/reason_mix.py --input before/returns.csv --platform amazon --sales before/sku-sales.csv --out after/reason-m
Reviews
Only verified buyers count; we do not edit, delete or pad them.This pack just launched and nobody has reviewed it yet. Once real buyers do, the stars, text and 1–5 distribution appear here; until then the rating shows an em dash instead of a made-up score.
Changelog
Updates are free while subscribed; a buyout includes later versions of this skillHow to buy
Subscribe or buy outright — a buyout means no more subscription feesMonthly
Try it for a month first
- ✓ “退货降本包” — all files, samples and install instructions
- ✓ Every skill unlocked while subscribed (including new releases)
- ✓ Commercial licence: client work allowed, unlimited deliveries
- ✓ Version updates pushed automatically, no repurchase
- ✓ 1:1 support on Telegram, replies within 12 hours on business days
- ✓ Cancel anytime, no next-period charge
Yearly
Pay for a year; prices are never converted to a monthly figure.
- ✓ “退货降本包” — all files, samples and install instructions
- ✓ Every skill unlocked while subscribed (including new releases)
- ✓ Commercial licence: client work allowed, unlimited deliveries
- ✓ Version updates pushed automatically, no repurchase
- ✓ 1:1 support on Telegram, replies within 12 hours on business days
- ✓ Cancel anytime, no next-period charge
Lifetime
The founder price rises as the skill library grows.
- ✓ “退货降本包” — all files, samples and install instructions
- ✓ Every skill unlocked while subscribed (including new releases)
- ✓ Commercial licence: client work allowed, unlimited deliveries
- ✓ Version updates pushed automatically, no repurchase
- ✓ 1:1 support on Telegram, replies within 12 hours on business days
- ✓ No subscription or renewals — one payment, forever
No price conversion: yearly plans only show the saving, never a per-month figure. A buyout is one payment with no renewals.