← 返回 Skills 市场
harrylabsj

Cheap

作者 haidong · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ✓ 安全检测通过
209
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install cheap
功能描述
Find the cheapest meaningful visible price for a user-provided product name across major Chinese shopping platforms such as Taobao, Tmall, JD.com, Pinduoduo,...
使用说明 (SKILL.md)

Cheap

Find the cheapest meaningful visible price for a product across Chinese e-commerce platforms.

Use this skill when the user provides a product name and wants the lowest current visible price, the cheapest platform, or a quick cross-platform price comparison.

This skill operates in two modes:

  • Mode 1: Price-accessible mode — enough platform price evidence is accessible to support a meaningful comparison.
  • Mode 2: Price-limited mode — platform access, product matching, or price visibility is too weak for a strong cheapest-price claim.

Read these references as needed:

  • references/product-matching.md when deciding whether two listings are likely the same product
  • references/platform-playbook.md when searching different Chinese shopping platforms
  • references/price-judgment.md when deciding which price counts as the meaningful cheapest price
  • references/output-patterns.md when preparing the final answer
  • references/failure-modes.md when product matching, pricing, or platform access is weak
  • references/examples.md when examples help calibrate the result format

Workflow

  1. Identify the product.

    • Accept a product name.
    • If the product is too broad or ambiguous, ask one short clarifying question for the most decision-relevant detail, such as brand, model, size, or key variant.
  2. Check price accessibility.

    • Search across relevant Chinese platforms.
    • Decide which mode applies:
      • Mode 1 if enough visible price evidence is accessible and comparable.
      • Mode 2 if platform access, listing clarity, or price conditions are too weak for a strong cheapest-price claim.
  3. If Mode 1, compare meaningful prices.

    • Prefer Taobao, Tmall, JD.com, Pinduoduo, and other clearly relevant platforms.
    • Use only currently visible price evidence.
    • Match comparable listings by brand, model, specs, quantity, version, and bundle contents.
    • Distinguish straightforward visible price from coupon-only, member-only, deposit/pre-sale, bundle-distorted, or variant-mismatched prices.
  4. If Mode 2, do not fake precision.

    • State that the currently accessible evidence is too weak for a strong cheapest-price claim.
    • Explain whether the issue is platform access, weak product matching, or distorted price conditions.
    • Give only a limited conclusion when still useful.
  5. Give the result. Cover:

    • cheapest platform or listing found when supportable
    • visible price evidence actually available
    • comparable alternatives when supportable
    • matching confidence
    • any caveats about variants, bundles, coupons, seller trust, or access limits

Output

Use this structure unless the user asks for something else.

Mode 1 output

Use when enough price evidence is accessible.

Cheapest Found

State the lowest meaningful visible price found and on which platform.

Comparable Options

List the main comparable prices found on other platforms.

Match Confidence

State whether the compared listings look like the same product:

  • High
  • Medium
  • Low

Caveats

State any important risks, such as:

  • unclear variant match
  • bundle differences
  • coupon or membership pricing
  • deposit or pre-sale pricing
  • seller quality concerns
  • incomplete platform access

Final Advice

Give a direct buying suggestion in plain language.

Mode 2 output

Use when accessible evidence is too weak for a strong lowest-price claim.

Current Best Signal

State the weakest still-usable price clue, if any.

Why No Strong Cheapest Claim

Explain whether the issue is:

  • platform access limits
  • weak listing comparability
  • unclear price conditions
  • incomplete platform coverage

What Remains Unverified

State what cannot yet be confirmed.

Final Advice

Give a cautious conclusion without pretending a definitive lowest price was found.

Quality bar

Do:

  • compare like-for-like listings
  • say when the cheapest visible option may not be the best option
  • distinguish real low price from misleading low price
  • be explicit when matching confidence is weak

Do not:

  • compare different variants as if they were the same item
  • treat coupon-only or pre-sale prices as normal prices without warning
  • claim full-market coverage when platform access is incomplete
  • invent prices or listing details

Limitation handling

If the product is too ambiguous:

  • ask one short clarifying question

If platform access is incomplete:

  • switch to Mode 2 unless a narrower Mode 1 comparison is still supportable
  • say which platforms were actually checked
  • avoid claiming a definitive all-platform lowest price
  • do not present a partial platform scan as a full-market cheapest-price result

If listing comparability is weak:

  • lower confidence
  • explain what may not match cleanly

If the visible lowest number depends on unclear conditions:

  • do not present it as the straightforward cheapest option without warning
安全使用建议
This skill appears internally consistent and low-risk: it asks for a product name and returns a cautious cross-platform price comparison using currently visible prices. Before installing, consider: (1) the agent will need web/network access to fetch platform listings — some Chinese platforms may require region-specific pages or logins, which can push the skill into Mode 2; (2) the skill does not perform purchases or need your credentials, so do not provide payment/account secrets to it; (3) the skill's results depend on live site scraping or queries (not included here), so verify any price it reports before buying because prices and promotions change quickly. If you expect the agent to log in to marketplaces or make purchases, do not install without reviewing the exact runtime browsing/integration behavior.
功能分析
Type: OpenClaw Skill Name: cheap Version: 0.1.0 The skill bundle contains instructions for an AI agent to perform price comparisons across Chinese e-commerce platforms (Taobao, JD.com, etc.). The logic focuses on data accuracy, product matching, and transparent reporting of limitations (Mode 1 vs. Mode 2). There is no executable code, no evidence of data exfiltration, and no malicious prompt injection; the instructions in SKILL.md and the references/ directory are strictly aligned with the stated purpose of finding the best prices.
能力评估
Purpose & Capability
The name and description match the runtime instructions: the skill is explicitly a cross-platform price lookup and comparison tool for Chinese e‑commerce sites. It does not request unrelated binaries, credentials, or configuration, and the references and workflow focus on product matching and visible-price judgment, which are appropriate for the stated purpose.
Instruction Scope
SKILL.md gives high-level, constrained runtime instructions (identify product, check platform prices, decide Mode 1/Mode 2, present results with caveats). It does not instruct the agent to read local files, environment variables, or other system state, nor does it instruct exfiltration to hidden endpoints. The instructions are deliberately cautious about conditional prices and about not inventing data.
Install Mechanism
There is no install spec and no code files; this instruction-only skill writes nothing to disk and does not pull external binaries or archives.
Credentials
The skill requires no environment variables, credentials, or config paths. All declared requirements are empty and proportionate to an instruction-only lookup/analysis skill.
Persistence & Privilege
Flags show always:false and default model invocation settings. The skill does not request persistent or elevated privileges, does not modify other skills, and has no install-time persistence behavior.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install cheap
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /cheap 触发
  4. 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
v0.1.0
Initial release: dual-mode Chinese e-commerce cheapest-price finder with meaningful price comparison, product matching, and price-limited fallback.
元数据
Slug cheap
版本 0.1.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Cheap 是什么?

Find the cheapest meaningful visible price for a user-provided product name across major Chinese shopping platforms such as Taobao, Tmall, JD.com, Pinduoduo,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 209 次。

如何安装 Cheap?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install cheap」即可一键安装,无需额外配置。

Cheap 是免费的吗?

是的,Cheap 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Cheap 支持哪些平台?

Cheap 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Cheap?

由 haidong(@harrylabsj)开发并维护,当前版本 v0.1.0。

💬 留言讨论