← 返回 Skills 市场
249
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install abandoned-checkout-monitor
功能描述
Deep cart-to-checkout funnel monitoring, abnormal friction detection, and multi-touch recovery playbooks for e-commerce. Use this skill whenever the user men...
使用说明 (SKILL.md)
Abandoned Checkout Monitor
You are a cart → checkout → payment diagnostician and recovery advisor. Your goal is to turn live cart behavior → friction detection → multi-touch recovery into an actionable full playbook, not scattered tips.
Mandatory full playbook (pushy policy)
Even if the user only asks "why no orders," "sales are slow," or "is our conversion broken" — as long as the topic is orders, checkout, or abandonment — you must still deliver all three blocks below (not a one-line answer):
- Checkout UI friction — checklist (fields, steps, trust, shipping disclosure, mobile) plus store-specific hypotheses.
- Payment gateway troubleshooting — self-serve steps aligned to common platforms (logs, test orders, region/currency, 3DS, webhooks, sandbox vs live).
- Three-email recovery sequence — Email 1 (gentle nudge + help), Email 2 (remove barriers + optional small incentive), Email 3 (last chance + human escalation); each with subject line A/B and body skeletons.
When data is missing, label assumptions and state what to instrument (events, funnel, payment error codes) to validate.
When NOT to use this skill (should-not-trigger)
- Only stock checks, whether a SKU is in stock, restock timing.
- Only a single order’s status, tracking number, or line-item export.
- In those cases, answer briefly; do not force the long template. If the user extends to "many people can't pay" or "checkout is broken," switch to the full playbook.
Gather context (infer from the thread first; ask only what’s missing)
- Platform (Shopify, WooCommerce, custom, etc.) and primary markets / currency.
- Checkout conversion or funnel: add to cart → begin checkout → purchase (if known).
- Whether certain regions or lanes have unusually high shipping; AOV bands and high-AOV SKUs.
- Payment methods (Stripe, PayPal, local wallets, etc.) and recent errors or chargebacks.
- Existing abandoned-cart email / SMS / retargeting; compliance (unsubscribe, frequency).
For deeper checklists, read references/abandonment_playbook.md when needed.
Success output: required structured master table
For every full response about abandonment, checkout drop-off, or recovery, include this Markdown table (at least 4 rows, spanning different drop-off points):
| Drop-off node | Likely cause (hypothesis) | A/B copy to test |
|---|---|---|
| (e.g. leave on cart page) | (e.g. shipping not shown early, free-shipping threshold unclear) | (e.g. A "You're $X from free shipping" vs B "This order qualifies for free shipping when…") |
| (e.g. after address on checkout) | (e.g. delivery time too long, no pickup option) | … |
| (e.g. payment step fail / back) | (e.g. 3DS fail, gateway timeout) | … |
| (e.g. high-AOV add-to-cart, no pay) | (e.g. trust, installments, returns clarity) | … |
Column meanings:
- Drop-off node: funnel step or event name (align to your platform’s events).
- Likely cause (hypothesis): separate "needs data" vs "common prior"; avoid vague fluff.
- A/B copy to test: testable copy or module pairs with a clear hypothesis (e.g. lift begin-checkout rate).
Beyond the table, include per the pushy policy: checkout UI friction, payment troubleshooting, three-email scripts (as subsections).
Recommended report outline (full playbook)
- Funnel snapshot — if data exists; otherwise define metrics and formulas to collect.
- Structured master table — required as above.
- Checkout UI friction — by module (form, shipping, trust, mobile).
- Payment gateway troubleshooting — step-by-step checklist.
- Three-email recovery scripts — subject A/B + bodies.
- Monitoring and next steps — event naming, review cadence.
How this skill fits with others
- Pure return rate / refunds → use a returns-focused skill.
- Pure site-wide CRO / homepage → use a CRO audit skill.
- This skill focuses on last-mile checkout, payment failure / shipping shock, and recovery outreach.
安全使用建议
This skill appears internally consistent and low-risk: it only contains instructions and reference docs to generate checkout-diagnostic playbooks, and it asks for no credentials or installs. Before installing, consider: (1) the skill is designed to always produce a long, structured playbook (table + checkout friction checklist + gateway troubleshooting + three-email sequence) even for vague questions — if you prefer short answers for simple inventory or order-status queries, do not rely on this skill for those. (2) The skill may ask you to supply funnel metrics, payment error codes, or sample logs to validate hypotheses — avoid pasting raw payment logs, full PII, or admin credentials; provide aggregated or anonymized metrics instead. (3) Recovery-email drafts may implicate legal/compliance requirements (CAN-SPAM, GDPR consent, local rules) — review any outreach plans with your legal/compliance team before sending. (4) If you want the agent to act autonomously across systems or to fetch logs from your platform, do not provide admin keys or live credentials to the agent; instead, extract and share only the minimally necessary aggregated data. Overall: technically benign and coherent; the primary non-security concern is its mandatory verbose output model and how much data you choose to share when following its instrumenting requests.
功能分析
Type: OpenClaw Skill
Name: abandoned-checkout-monitor
Version: 0.1.1
The 'abandoned-checkout-monitor' skill is a diagnostic and advisory tool for e-commerce checkout optimization. It uses structured Markdown instructions (SKILL.md and references/abandonment_playbook.md) to guide the AI agent in providing checkout friction audits, payment troubleshooting, and email recovery templates. The skill contains no executable code, does not request sensitive system access (like environment variables or SSH keys), and lacks any indicators of data exfiltration or malicious intent.
能力评估
Purpose & Capability
The skill's name, description, and SKILL.md align: it is a diagnostic + playbook generator for cart→checkout abandonment. It asks for platform, markets/currency, funnel metrics, payment methods, and existing outreach — all relevant to the stated goal. No unrelated binaries, env vars, or config paths are requested.
Instruction Scope
The SKILL.md is prescriptive (requires a multi-block 'full playbook' and a structured master table with at least four rows). It instructs the agent to read the included reference playbook and to infer context from the conversation, asking only missing questions. It does not instruct reading system files, environment variables, or contacting external endpoints. Note: the policy is 'pushy' — the skill will produce long, structured output even for vague merchant questions, which is a design choice (not a security issue) but may produce lengthy disclosures if a user supplies logs or data.
Install Mechanism
No install spec and no code files that execute — instruction-only. This is low-risk: nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. All requested context (platform, metrics, payment methods, existing outreach) is appropriate for its function.
Persistence & Privilege
always:false (no forced inclusion). Model invocation is permitted (default) but not combined with any credential requests or system-level changes. The skill does not request persistent system modifications or access to other skills' configs.
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 abandoned-checkout-monitor - 安装完成后，直接呼叫该 Skill 的名称或使用
/abandoned-checkout-monitor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
v0.1.1
Version 0.1.1 of abandoned-checkout-monitor is an English translation and adaptation of the original Chinese skill documentation.
- All documentation and guidelines converted fully to English, with context and examples rewritten for global e-commerce audiences.
- Trigger conditions and exclusions updated to use concise English phrases and broader applicability.
- Mandatory output structure, report outlines, and example tables retained; field names and explanations presented in English.
- Tone and instructions clarified for users familiar with international e-commerce platforms and best practices.
v0.1.0
abandoned-checkout-monitor v0.1.0
- Initial release with comprehensive guidance for diagnosing and recovering abandoned checkouts in e-commerce.
- Introduces a “Pushy” strategy: always deliver a full end-to-end response including UI friction analysis, payment gateway troubleshooting, and a 3-step recovery email sequence.
- Requires structured output with a markdown table mapping dropout points to likely causes and A/B messaging ideas.
- Defines clear triggers and exclusions to ensure skill is used only for checkout-related issues, not general order or inventory queries.
- Includes detailed instructions for context gathering and report structuring for recovery workflows.
元数据
常见问题
Abandoned Checkout Monitor 是什么？
Deep cart-to-checkout funnel monitoring, abnormal friction detection, and multi-touch recovery playbooks for e-commerce. Use this skill whenever the user men... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 249 次。
如何安装 Abandoned Checkout Monitor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install abandoned-checkout-monitor」即可一键安装，无需额外配置。
Abandoned Checkout Monitor 是免费的吗？
是的，Abandoned Checkout Monitor 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Abandoned Checkout Monitor 支持哪些平台？
Abandoned Checkout Monitor 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 Abandoned Checkout Monitor？
由 RIJOY-AI（@rijoyai）开发并维护，当前版本 v0.1.1。
推荐 Skills