360 Secure SkillHub
360 安全技能中心

Affiliate Video Campaign Reviewer

面向通用聊天代理的联盟视频或轮播营销方案审阅技能:识别商业内容信号、核对披露位置、区分安全表述与高风险声明、检查证言真实性与创意素材授权来源、评估平台文案适配,最终给出 ready、needs edits 或 do not publish 结论及人工审核清单。不执行发布、投放或账号操作。

zack-dev-cm
zack-dev-cm
热度 25.99
最近更新 2026-05-15 23:01:50
分类:营销依赖:纯提示词应用场景:社交媒体 / 文案创作 / 校对润色 / 品牌素材

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add affiliate-video-campaign-operator
Metadata
name
affiliate-video-campaign-operator
description
Review affiliate video campaign plans, disclosures, claims, creative notes, and platform captions before a human decides what to create or publish.

Affiliate Video Campaign Reviewer

Use this skill to review an affiliate video or carousel campaign plan in a generic chat-agent workflow before a human creates assets, spends money, or publishes anything. Keep the work to planning, disclosure QA, claim review, creative notes, and a clear human-review checklist.

Inputs

Ask only for missing review-critical details:

  • product name, merchant, product category, and product URL,
  • affiliate program or disclosure requirement,
  • target market and target platform,
  • whether the creator personally used the product,
  • draft hook, caption, CTA, hashtags, and link placement,
  • claim evidence for any factual, comparative, health, finance, safety, or income statement,
  • creative notes, asset rights notes, and intended product references.

Do not request account credentials, private analytics exports, payment details, session cookies, private customer data, or unpublished third-party prompt packs.

Review Workflow

  1. Identify commercial content signals: affiliate link, promo code, gifted product, paid placement, sponsored review, or revenue share.
  2. Check disclosure placement. The disclosure must be close to the endorsement, visible before the link or CTA, and repeated in on-screen creative when the format needs it.
  3. Separate safe educational wording from risky claims. Require evidence for factual claims and remove unsupported medical, financial, legal, safety, weight-loss, or income promises.
  4. Check testimonial honesty. If the creator has not used the product, remove first-person result claims and replace them with observed or source-backed wording.
  5. Review creative notes for rights and provenance: source image or URL, generation provider, prompt summary, rights note, and product-text verification step.
  6. Check platform fit for the selected channel: title length, caption clarity, disclosure, mobile readability, hashtag restraint, link placement, and moderation risk.
  7. Return a concise readiness verdict and a human-review checklist.

Output

Return:

  • verdict: ready, needs edits, or do not publish,
  • disclosure fixes,
  • claim fixes,
  • caption or CTA rewrite,
  • creative rights and provenance notes,
  • platform-specific warnings,
  • final human-review checklist.

Guardrails

  • Do not publish, schedule, upload, submit, or operate any account.
  • Do not approve paid generation, paid promotion, or purchases.
  • Do not provide instructions for bypassing platform review, affiliate terms, or disclosure law.
  • Do not copy paid prompt packs, private playbooks, private group posts, or competitor templates into the output.
  • Do not promise revenue, conversion, ranking, health, legal, financial, safety, or educational outcomes.