ab-test-plan
Design a statistically rigorous A/B or multivariate test plan — If/Then/Because hypothesis, control and variant specs, required sample size per variant…
Generate or critique video packaging — title + thumbnail-text pairs where the title carries context and keywords, the thumbnail text carries the tension, and any word echoed between them is rejected as wasted real estate. Generate mode delivers 3 pairs tagged by discovery intent
$ npx -y skills add indranilbanerjee/digital-marketing-pro --skill video-packaging --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/video-packagingContext preview
The summary Claude sees to decide when to auto-load this skill.
Generate or critique video packaging — title + thumbnail-text pairs where the title carries context and keywords, the thumbnail text carries the tension, and any word echoed between them is rejected as wasted real estate. Generate mode delivers 3 pairs tagged by discovery intent
name: video-packaging description: "Generate or critique video packaging — title + thumbnail-text pairs where the title carries context and keywords, the thumbnail text carries the tension, and any word echoed between them is rejected as wasted real estate. Generate mode delivers 3 pairs tagged by discovery intent (search vs browse) with a recommendation and A/B note; critique mode returns a PASS/FIX/FAIL verdict with 3 fixed pairs. Triggers on \"/digital-marketing-pro:video-packaging\", \"title for this video\", \"thumbnail text ideas\", \"why is no one clicking this video\", \"critique this title\", \"package this video\". Pairs with /digital-marketing-pro:video-script, pulls real query phrasing from /digital-marketing-pro:keyword-research for search-intent titles, and gates title claims through /digital-marketing-pro:check before anything ships." argument-hint: "[brand-name] [--topic <video topic>] [--critique \"<existing title> / <thumb text>\"]" user-invocable: true
Packaging decides whether a video gets watched; the content only decides whether it deserved to be. On every video surface the viewer sees two elements at once — the title and the thumbnail text — and the craft is that **they must do different jobs**:
way people search and the way the platform's systems read it.
the "wait, what?" that a human reacts to in half a second.
title bought nothing with the scarcest pixels the brand owns.
A video idea earns discovery one of two ways, and the packaging differs:
| Intent | How it is found | Packaging consequence | |---|---|---| | **Search** | Typed queries; ranks over months | Title leads with the query phrasing (front-load the keyword); thumb adds the differentiator vs the other results | | **Browse** | Suggested/home feeds; spikes on click-through | Title can spend more of its length on intrigue; thumb carries the pattern-break | | **Both** | Rare; usually a search topic with a browse-worthy angle | Package for search, let the thumb do the browse work |
**No search or browse logic → not a video idea yet.** Route it back through topic development before spending packaging effort on it. For search-intent videos, pull the real query phrasing from /digital-marketing-pro:keyword-research rather than guessing it.
Input: the topic or concept, the brand, target platform, intent if known.
# Video packaging — [topic] **Discovery intent:** Search / Browse / Both — [the reasoning in one line] ## Pair 1 — [angle: curiosity / benefit / identity] **Title:** [≤60 chars where possible — what survives truncation comes first] **Thumb text:** [1-3 words] **Why they pair:** [the different jobs, and how they combine into one click decision] ## Pair 2 — [different angle] ## Pair 3 — [different angle] ## Recommendation [Which pair leads and why — grounded in the intent tag, not taste] ## A/B note [Which single element to vary if testing — never both at once]
Input: an existing title and/or thumbnail text (works from a screenshot description too).
# Packaging critique — [the draft] ## Verdict: PASS / FIX / FAIL ## The pairing check [Do title and thumb do different jobs? Quote any echoed words — each is the finding, verbatim.] ## Working [What earns its place] ## Not working [Specific, quotable problems — vague titles, buried keywords, thumb text restating the title, curiosity with no clarity] ## 3 fixed pairs [Revised packages, same content, same claims — packaging never promises what the video does not deliver]
countable core rule — check it word by word, not by impression.
matters sits before the truncation point.
viewer gets fails — curiosity-gap packaging that the content cannot pay off is churn, and on regulated brands it is a compliance finding.
same brand guardrails as any other copy; run /digital-marketing-pro:check before anything ships.
option is a decision already made; the client gets choices with reasoning.
video will be found — and search and browse reward different structures.
its thumbnail-concept step follows the same pairing rule
search-intent titles
Your agency just signed a 50-brand client. The previous agency left no playbook. Three brands are bleeding budget, two have stale positioning, one is launching in a regulated jurisdiction next month. Where do you start?
Repo: indranilbanerjee/digital-marketing-pro
Design a statistically rigorous A/B or multivariate test plan — If/Then/Because hypothesis, control and variant specs, required sample size per variant…
Generate 3-5 ad copy variations per platform — headlines, descriptions, and CTAs formatted to Google, Meta, LinkedIn, TikTok, X, and Pinterest specs — each…
Walk through adding a custom MCP server integration to the plugin — searches npm for an existing MCP package (or scaffolds a custom server from the plugin's…
Audit how a brand appears across the 6 canonical AI answer surfaces — ChatGPT, Perplexity, Google AI Mode, AI Overviews, Gemini, Copilot — probing 10-25…
Strategy module for Answer Engine / Generative Engine Optimization — audits AI visibility, restructures content for citation, runs entity-consistency checks…
Generate a portfolio-level dashboard across ALL client brands — per-client RAG health scores, campaign activity, budget pacing, aggregate KPIs, team…