AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Sealeap Baize Amazon Ad Conversion States

skill-xjli360-sealeap-amazon-skills-sealeap-baize-amazon-ad-conversion-states · by xjli360

Diagnose Amazon ad conversion that declines, stays weak, or fluctuates by separating placement expansion, price-value fit, query relevance, product-page differentiation, and market events. Use for 广告转化越来越差、一直不出单、转化忽高忽低、点击增加但订单不增. Do not use to make live campaign changes without approval.

— No reviews yet
0 installs
26 views
0.0% view→install

Install

$ agentstack add skill-xjli360-sealeap-amazon-skills-sealeap-baize-amazon-ad-conversion-states

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-xjli360-sealeap-amazon-skills-sealeap-baize-amazon-ad-conversion-states)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
● 16d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Sealeap Baize Amazon Ad Conversion States? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Amazon 广告转化状态诊断

目标

Diagnose Amazon ad conversion that declines, stays weak, or fluctuates by separating placement expansion, price-value fit, query relevance, product-page differentiation, and market events.

不可妥协的边界

  • 当前 Amazon 官方政策、账户资格、站点字段和一方数据优先于本 Skill 的经验框架。
  • 第三方数据一律标为估算或前台观测,不得写成 Amazon 一方事实。
  • 默认提供诊断或草案。写入前展示对象、旧值、新值、影响、停止线与回退,核对用户已有授权是否覆盖对象、动作与预算;范围已明确授权时继续执行并回读核验,只有未覆盖或扩大的范围才请求批准。
  • 一次实验只改变一个主要变量,并记录基线、样本、成功、停止和回退条件。
  • 不得复制来源材料或竞品表达;输出必须按当前任务重新组织并可由现有证据支撑。
  • 位置靠前不必然更好;必须用本 ASIN 的位置报告验证。
  • 竞品自然排名只能作为相关性线索,不能证明本品会转化。

先判断任务模式

  1. 诊断:读取现状、证据和缺口,不生成线上写入动作。
  2. 方案草案:输出可审核的结构、参数范围、实验和回退值。
  3. 执行准备:只生成待批准变更表或 API/控制台操作草案。
  4. 已批准执行:仅对用户在当前会话明确批准的对象和字段执行,并立即回读核验。

用户未指定时采用“诊断”。

开始前要拿到

  • marketplace、店铺、ASIN/SKU、广告类型和目标
  • 同口径的 Campaign、Targeting、Search Term、Placement 与 Advertised Product 报告
  • 售价、优惠、COGS、Amazon 费用、退款与目标贡献利润
  • 库存、Buy Box、Listing、评论和同期市场事件

缺失项必须标为 NEEDS_EVIDENCE;不得猜数字、补属性或把不同站点、ASIN、变体、币种和时间窗混在一起。

工作流

先读取 [references/playbook.md](references/playbook.md),确认该方法适用于当前对象。按以下顺序执行:

  1. 先把问题分成持续恶化、长期偏弱和短期波动三种状态,使用同口径时间窗比较。
  2. 若点击速度上升而订单不增,先检查位置和查询是否扩展到低意向流量。
  3. 若长期偏弱,依次核对价格与价值表达、关键词意图、评论基础和页面差异。
  4. 若短期波动,记录竞品促销、价格、库存、Buy Box 与季节事件,不把单日变化当趋势。
  5. 只选择一个主变量做小幅实验,并预先写明花费上限和恢复值。

最后做数据充分性检查,并把结论分成 FACT / ESTIMATE / HYPOTHESIS / UNKNOWN。若关键证据不足,状态写 HOLD。

第三方 MCP 数据

仅在自有数据不足且当前任务确实需要外部证据时,读取 [references/mcp-data-plan.md](references/mcp-data-plan.md),再使用 scripts/mcp_research.py。本 Skill 的外部取数目的:补充关键词、竞品自然排名和市场价格的第三方代理证据。

  • 先 doctor,再 search-tools 和 describe;工具名及参数以实时 tools/list 与 inputSchema 为准。
  • Token 只从环境变量读取。不得写入命令参数、URL、Skill、报告、日志或 Git。
  • tools/call 或 Actor 可能计费;先展示 Provider、工具、无密钥业务参数、预计成本与输出位置,核对已有授权覆盖后才加 --allow-cost;该标志不是费用上限。

必须交付的结果

  • 三类状态判定
  • 根因证据矩阵
  • 单变量实验卡
  • 待批准的最小调整
  • 数据范围、来源、采集时间、样本与限制。
  • 关键假设、待补证据、风险和不可确定项。
  • 若有动作:对象、旧值、新值、预期、停止条件、回退值与审批状态。

方案状态使用 READY FOR REVIEW / DRAFT / HOLD / STOP;如已执行,另行记录实际结果及回读证据。未得到明确批准时,不得声称已修改线上对象。

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.