skills.rest
techfundoffice
techfundofficeCOMMUNITY
Explore all skills by this author →
GitHub

product-pick-blurb

Defines the three-sentence editorial template for Amazon affiliate product pick blurbs.

Updated Jul 21, 2026
Search Tags:#prompt engineering#copywriting#content audit#affiliate marketing#seo articles#editorial template#product blurbs
One-click install
npx skills add https://github.com/techfundoffice/cats-seo-aiagent-staging --skill product-pick-blurb-techfundoffice
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: product-pick-blurb
Source: https://github.com/techfundoffice/cats-seo-aiagent-staging/tree/main/.claude/skills/product-pick-blurb
Command: npx skills add https://github.com/techfundoffice/cats-seo-aiagent-staging --skill product-pick-blurb-techfundoffice

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Affiliate articles on catsluvus.com render an "Our Top Picks" card where each product needs a consistent editorial blurb, but the AI writer can drift in structure, voice, and length without a documented standard. This Skill keeps the writer prompt in src/pipeline/writer.ts and the rendered HTML output in sync by codifying the exact blurb format. ## Core Features & Use Cases - Blurb Structure Spec: Enforces a three-sentence, 55-85 word format (feature sentence, tradeoff-plus-fit sentence, closing line) for every product pick. - Closing-Line Template: Mandates the literal prefix "Why we like this pick:" followed by a problem-benefit-buyer pattern of 15-30 words. - Voice and Failure-Mode Rules: Bans AI-slop vocabulary, em dashes, and keyword stuffing, and documents how to detect drift by grep-counting closing lines in published HTML. - Use Case: When tuning the Kimi K2.5 writer prompt or auditing published articles, apply this template so every pickReasons entry stays consistent across articles. ## Quick Start Ask the agent to audit the latest published article's product blurbs against the three-sentence structure and closing-line template defined in this skill.

Frequently Asked Questions about product-pick-blurb

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write consistent product pick blurbs for affiliate articles?▼

Use a fixed three-sentence structure of 55-85 words: a concrete feature sentence, a tradeoff-plus-fit sentence, and a closing line starting with "Why we like this pick:". The closing line follows a problem, benefit, ideal-buyer pattern in 15-30 words.

How to detect when AI-generated product blurbs drift from the template?▼

Grep-count the literal string "Why we like this pick:" in published HTML and compare it against picks multiplied by articles. A mismatch means the model omitted closing lines, since the renderer has no enforcement.

What vocabulary should be banned in AI product copywriting?▼

Ban AI-slop terms like delve, leverage, utilize, robust, comprehensive, pivotal, cutting-edge, game-changer, and revolutionize. Also prohibit em dashes, exclamation points, keyword stuffing, and restating the product name in the closing line.

Why do product blurbs get truncated in rendered articles?▼

The normalizePickReasons() function caps the reasoning field at 500 characters, which fits the 55-85 word budget with headroom. If clipping becomes visible, raise the cap rather than tightening the word budget.

When should the writer prompt be updated alongside this template?▼

Update buildArticlePrompt() in src/pipeline/writer.ts in the same commit whenever the blurb structure changes. The prompt must restate the three-sentence structure and closing-line template so the model has the rules in context.

Related Skills

Explore More Content & Communication Skills→
hanzlahabibhanzlahabib
2

rcode-seo-site-builder

Discover and structure profitable affiliate SEO website opportunities from niche discovery to Next.js site launch.

Community
Advanced
anxun-agentanxun-agent

seo-article-html-generator

Generate SEO-optimized HTML articles with inline CSS from structured data.

Community
Intermediate
reflusterrefluster
1

article-health

Audits article corpora by comparing gh-pages publications with Notion content to detect truncation and drift.

Community
Intermediate
ajayatwal1105-emersonajayatwal1105-emerson

message-test

Predict message variant performance on synthetic audience panels.

Community
Advanced