Agent Skills › gooseworks-ai/goose-skills › run-paid-campaigns-health-gate

run-paid-campaigns-health-gate

GitHub

检查广告健康度，验证转化追踪、落地页、库存、审核及账户状态，将失败项转为修复建议，确保广告投放安全。

skills/marketing/run-paid-campaigns-health-gate/SKILL.md gooseworks-ai/goose-skills

触发场景

需要验证广告是否具备投放条件时检查现有或计划中广告的合规性与风险时

安装

npx skills add gooseworks-ai/goose-skills --skill run-paid-campaigns-health-gate -g -y

更多选项

非标准路径

npx skills add https://github.com/gooseworks-ai/goose-skills/tree/main/skills/marketing/run-paid-campaigns-health-gate -g -y

不安装直接使用

npx skills use gooseworks-ai/goose-skills@run-paid-campaigns-health-gate

指定 Agent (Claude Code)

npx skills add gooseworks-ai/goose-skills --skill run-paid-campaigns-health-gate -a claude-code -g -y

安装 repo 全部 skill

npx skills add gooseworks-ai/goose-skills --all -g -y

预览 repo 内 skill

npx skills add gooseworks-ai/goose-skills --list

SKILL.md

Frontmatter

{
    "name": "run-paid-campaigns-health-gate",
    "version": "0.1.2",
    "description": "Checks that ads are safe to spend on: conversion tracking recording, landing pages loading on mobile with the right product, price and offer, enough stock behind promoted products, no disapproved ads, and no account or payment stop. Marks each pass, fail or unknown and turns fails into fixes. Use alone when you already have the live or planned ads and their landing pages; otherwise run run-paid-campaigns."
}

Health gate (atom)

Load: ../run-paid-campaigns/references/launch-and-graduate.md (health gate section).

Inputs: the chosen platforms (from the brief); the latest analyze-performance read (reports/performance/) — its data-quality flags and the platform-claims ÷ store-orders ratio; with GooseWorks, meta_read push status for each id in state.md's meta_push_ids (per-creative failures, destinations; local mode: the connector's delivery and disapproval reads, ../run-paid-campaigns/references/local-mode.md); destination URLs of other live ads (the marketer's export); stock on promoted SKUs (marketer or product record); live offers and their end dates (brief's Offer section, the marketer); the marketer's ads connector, when there is one (disapprovals, delivery status, whether the conversion event is recording).

Do — in order, one line each, ✓ / ✗ / unknown:

Tracking: conversions tracking store orders at a steady ratio (from analyze-performance), the connector showing the conversion event recording, or the marketer confirms it fires. Lead-gen: lead source captured in the CRM. Apps: SDK/MMP postbacks, SKAN/AEM set up. Unknown on a launch → ask; do not launch blind.
Destination: loads on mobile, right product, price matches the ad, offer valid, UTMs present. Open the page when the host can browse; else ask.
Stock: no ad sends traffic to a SKU with under ~1 week of cover.
Disapprovals / failed creatives (push_status; the connector; the marketer's export).
Account: (Meta chosen only) publishable Page exists; every platform: no known payment failure or spending-limit stop.

Output: the proposal's health line, plus fix moves (pause the affected Goose ad, a card for non-Goose objects, a written tracking fix). Any ✗ blocks launches and scale moves for that destination this week. Only a tracking break makes the window's numbers suspect; low stock or a disapproval does not invalidate past results.