Claude Skill

product-review

Product-review editor for YouTube and social: verdict teased up front, what it is and the price, readable spec graphics, hands-on and macro B-roll over the talking, pros and cons cards, a final score, and a sponsorship / free-sample disclosure card, in 16:9 plus a 9:16 verdict cu

LLM Mart · 0 points · 0 views 0 listing impressions 0 install-command copies
Download viprasol-tech-claude-video-editing-skills-plugins_video-studio_skills_product-review-0293497.zip · 3 KB

Install

skills CLI npx skills add https://github.com/Viprasol-Tech/claude-video-editing-skills/tree/main/plugins/video-studio/skills/product-review
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install viprasol-tech-claude-video-editing-skills@llmmart
Git git clone https://github.com/Viprasol-Tech/claude-video-editing-skills.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole viprasol-tech/claude-video-editing-skills collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Product Review

Talking head plus clean product B-roll, structured so the verdict is clear and the trust is earned.

  • Best for: tech, gear, beauty, kitchen, car, app reviews; unboxing + review
  • Inputs: the talking A-roll, product B-roll (macro, hands-on, in use) in a folder named by what each shot shows, specs sheet, price, sponsorship status
  • Default skin: chapter-pro · alternatives product-spotlight (product-led), data-story (spec-heavy), split-proof (comparisons)
  • Formats: 16:9 1920x1080 master, 9:16 verdict cutdown (30-60 s), chapters, thumbnail stills
  • Duration target: 4-15 min; genre preset product-review

Before you start

  • Needs video-studio-core in the same skills folder. CORE = video-studio-core, SKILLS = the folder holding both.
  • First time: python CORE/scripts/check_env.py. Preset: python CORE/scripts/editor.py genre product-review.
  • The --genre preset sets autocut's pause cut (0.45 s) and run.py respects it; pass --max-gap only to override.
  • Ethics: ask whether the product was paid for, sent free, or sponsored. Any of the last two needs a clear disclosure on screen near the start and in the description. The score and the cons are the reviewer's real opinion: never edit a negative into a positive.

How a pro cuts this

beat % what happens
verdict tease 0-5 "This is the best X under $500 - with one big catch" + hero shot
disclosure + what it is 5-13 sponsor/free-sample card, name, price, who it's for
specs 13-25 the 3-6 specs that matter, as a card or bars, not a spec dump
hands-on 25-60 real use, macro B-roll every claim, comparisons
pros / cons 60-84 separate cards, balanced
verdict + score 84-96 who should buy, who shouldn't, score
end screen 96-100 next review
  • Pacing: ASL 3-6 s; B-roll over at least half of the talking; a change every 4-5 s.
  • Cut rules: jump cuts on the presenter are fine; cover the ones mid-explanation with the product shot of what is being said. B-roll must show the thing being described (not decoration).
  • J/L cuts: L-cut the presenter's line over the macro shot; J-cut the product sound (click, hum) before its shot.
  • Transitions: cuts; a whip or slide only on section changes. Speed ramp on one hero reveal.
  • Music/SFX: clean bed 24 LU under; UI ticks on spec items; a soft impact on the score.
  • Colour: product colour must be true: correct, then clean-commercial 0.4. Skin on the presenter stays natural.
  • Captions: sidecar SRT; on-screen text for specs, prices (with date), pros/cons and the score.
  • Client judges: trust, a clear verdict, clean product B-roll, readable spec graphics.
  • Common mistakes: spec dumps read off a sheet, a verdict only at minute 12, prices without a date, B-roll of the wrong model/colour, missing disclosure, cons edited out.

Make the edit

python CORE/scripts/run.py review.mp4 --style SKILLS/chapter-pro --genre product-review --smart-cuts \
  --retakes --broll broll/ --model small --prompt "Product Name, brand terms" --work rev --stop-after plan
python CORE/scripts/story.py outline --work rev --genre product-review      # find the verdict line for the tease
python CORE/scripts/editor.py review rev/plan.json --genre product-review --fix
python CORE/scripts/render.py rev/plan.json --frames 2,30,120,300
python CORE/scripts/render.py rev/plan.json
python CORE/scripts/finish.py rev/plan.json --platform youtube

If the verdict line was said at the end, cold-open on it with story.py reorder --work rev --order "s58-s59,s1-s57,s60-s70" and rebuild via run.py - --timeline rev/story_timeline.json ... (this is normal entertainment reordering; keep the line's meaning intact).

Refine the plan

  • Disclosure: {"type": "chip", "text": "Sponsored by BRAND", "start": 6.0, "end": 11.0, "x": 60, "y": 60, "anchor": "top-left"} (or "Sent free by BRAND - no money, no review approval").
  • Product + price: {"type": "card", "kicker": "Reviewed", "title": "Model X2", "body": "$499 (price on 17 Nov 2026)", "image": "{image}"} - pass --image product.png to run.py.
  • Specs: {"type": "bars", "title": "Battery (hours)", "items": [{"label": "X2", "value": 31, "highlight": true}, {"label": "Rival", "value": 24}]} or a card with items and at times.
  • Pros / cons: two checklist elements ("Pros", "Cons") with items timed to when they are said.
  • Score: {"type": "counter", "from": 0, "to": 8.5, "decimals": 1, "suffix": "/10", "label": "Verdict"} + "sfx": [{"t": 512.0, "kind": "impact", "gain": 0.5}].
  • Comparisons: compare (before/after images) or split with two panes.
  • Look/sound: "color": {"look": "clean-commercial", "intensity": 0.4}, "music": {"src": "bed.mp3", "under": 24, "duck": "speech-aware"}.
  • Chapters: top-level "chapters" for verdict, specs, hands-on, pros, cons, verdict.

Deliver

python CORE/scripts/deliver.py rev/plan.json --to youtube --out-dir rev/deliver
python CORE/scripts/qc.py rev/deliver/final_youtube.mp4 --plan rev/plan.json --platform youtube
  • 9:16 verdict cutdown: retention-reel on the verdict section (prep.py review.mp4 --work rev-short --start S --end E).
  • Description: disclosure text first, then chapters.txt; check thumbnails show the actual product.

Check before delivering

  • Disclosure on screen in the first minute and in the description whenever money or free product was involved.
  • Every spec, price and model name verified; price dated.
  • B-roll matches the model/colour talked about; product colour accurate on a contact sheet.
  • Pros and cons both present; score matches what was said.
  • qc.py passes; tell the user structure, chapters and files.

Example requests

  • "Edit my phone review - tease the verdict first, spec cards, pros and cons, score at the end."
  • "Sponsored headphone review, add the disclosure and macro B-roll over the talking."
  • "Unboxing + review into a 10-minute video and a 45-second verdict Short."
Files (claude-video-editing-skills)
  • SKILL.md 6.6 KB
    ---
    name: product-review
    description: "Product-review editor for YouTube and social: verdict teased up front, what it is and the price, readable spec graphics, hands-on and macro B-roll over the talking, pros and cons cards, a final score, and a sponsorship / free-sample disclosure card, in 16:9 plus a 9:16 verdict cutdown. Use when the user says: edit my review, tech review, product review video, unboxing and review, pros and cons video, gadget review, rate this product, sponsored review. Reviews must be honest and disclosed (FTC / ASCI). Default skin product-spotlight over a talking head, or any Video Studio skin."
    license: MIT
    ---
    
    # Product Review
    
    Talking head plus clean product B-roll, structured so the verdict is clear and the trust is earned.
    
    - **Best for:** tech, gear, beauty, kitchen, car, app reviews; unboxing + review
    - **Inputs:** the talking A-roll, product B-roll (macro, hands-on, in use) in a folder named by what each shot shows, specs sheet, price, sponsorship status
    - **Default skin:** `chapter-pro` · alternatives `product-spotlight` (product-led), `data-story` (spec-heavy), `split-proof` (comparisons)
    - **Formats:** 16:9 1920x1080 master, 9:16 verdict cutdown (30-60 s), chapters, thumbnail stills
    - **Duration target:** 4-15 min; genre preset `product-review`
    
    ## Before you start
    
    - Needs **video-studio-core** in the same skills folder. `CORE` = `video-studio-core`, `SKILLS` = the folder holding both.
    - First time: `python CORE/scripts/check_env.py`. Preset: `python CORE/scripts/editor.py genre product-review`.
    - The `--genre` preset sets autocut's pause cut (0.45 s) and run.py respects it; pass `--max-gap` only to override.
    - **Ethics:** ask whether the product was paid for, sent free, or sponsored. Any of the last two needs a clear disclosure on screen near the start and in the description. The score and the cons are the reviewer's real opinion: never edit a negative into a positive.
    
    ## How a pro cuts this
    
    | beat | % | what happens |
    | --- | --- | --- |
    | verdict tease | 0-5 | "This is the best X under $500 - with one big catch" + hero shot |
    | disclosure + what it is | 5-13 | sponsor/free-sample card, name, price, who it's for |
    | specs | 13-25 | the 3-6 specs that matter, as a card or bars, not a spec dump |
    | hands-on | 25-60 | real use, macro B-roll every claim, comparisons |
    | pros / cons | 60-84 | separate cards, balanced |
    | verdict + score | 84-96 | who should buy, who shouldn't, score |
    | end screen | 96-100 | next review |
    
    - **Pacing:** ASL 3-6 s; B-roll over at least half of the talking; a change every 4-5 s.
    - **Cut rules:** jump cuts on the presenter are fine; cover the ones mid-explanation with the product shot of what is being said. B-roll must show *the thing being described* (not decoration).
    - **J/L cuts:** L-cut the presenter's line over the macro shot; J-cut the product sound (click, hum) before its shot.
    - **Transitions:** cuts; a whip or slide only on section changes. Speed ramp on one hero reveal.
    - **Music/SFX:** clean bed 24 LU under; UI ticks on spec items; a soft impact on the score.
    - **Colour:** product colour must be true: correct, then `clean-commercial` 0.4. Skin on the presenter stays natural.
    - **Captions:** sidecar SRT; on-screen text for specs, prices (with date), pros/cons and the score.
    - **Client judges:** trust, a clear verdict, clean product B-roll, readable spec graphics.
    - **Common mistakes:** spec dumps read off a sheet, a verdict only at minute 12, prices without a date, B-roll of the wrong model/colour, missing disclosure, cons edited out.
    
    ## Make the edit
    
    ```bash
    python CORE/scripts/run.py review.mp4 --style SKILLS/chapter-pro --genre product-review --smart-cuts \
      --retakes --broll broll/ --model small --prompt "Product Name, brand terms" --work rev --stop-after plan
    python CORE/scripts/story.py outline --work rev --genre product-review      # find the verdict line for the tease
    python CORE/scripts/editor.py review rev/plan.json --genre product-review --fix
    python CORE/scripts/render.py rev/plan.json --frames 2,30,120,300
    python CORE/scripts/render.py rev/plan.json
    python CORE/scripts/finish.py rev/plan.json --platform youtube
    ```
    
    If the verdict line was said at the end, cold-open on it with `story.py reorder --work rev --order "s58-s59,s1-s57,s60-s70"` and rebuild via `run.py - --timeline rev/story_timeline.json ...` (this is normal entertainment reordering; keep the line's meaning intact).
    
    ## Refine the plan
    
    - **Disclosure:** `{"type": "chip", "text": "Sponsored by BRAND", "start": 6.0, "end": 11.0, "x": 60, "y": 60, "anchor": "top-left"}` (or "Sent free by BRAND - no money, no review approval").
    - **Product + price:** `{"type": "card", "kicker": "Reviewed", "title": "Model X2", "body": "$499 (price on 17 Nov 2026)", "image": "{image}"}` - pass `--image product.png` to run.py.
    - **Specs:** `{"type": "bars", "title": "Battery (hours)", "items": [{"label": "X2", "value": 31, "highlight": true}, {"label": "Rival", "value": 24}]}` or a `card` with `items` and `at` times.
    - **Pros / cons:** two `checklist` elements ("Pros", "Cons") with items timed to when they are said.
    - **Score:** `{"type": "counter", "from": 0, "to": 8.5, "decimals": 1, "suffix": "/10", "label": "Verdict"}` + `"sfx": [{"t": 512.0, "kind": "impact", "gain": 0.5}]`.
    - **Comparisons:** `compare` (before/after images) or `split` with two panes.
    - **Look/sound:** `"color": {"look": "clean-commercial", "intensity": 0.4}`, `"music": {"src": "bed.mp3", "under": 24, "duck": "speech-aware"}`.
    - **Chapters:** top-level `"chapters"` for verdict, specs, hands-on, pros, cons, verdict.
    
    ## Deliver
    
    ```bash
    python CORE/scripts/deliver.py rev/plan.json --to youtube --out-dir rev/deliver
    python CORE/scripts/qc.py rev/deliver/final_youtube.mp4 --plan rev/plan.json --platform youtube
    ```
    
    - 9:16 verdict cutdown: `retention-reel` on the verdict section (`prep.py review.mp4 --work rev-short --start S --end E`).
    - Description: disclosure text first, then `chapters.txt`; check thumbnails show the actual product.
    
    ## Check before delivering
    
    - Disclosure on screen in the first minute and in the description whenever money or free product was involved.
    - Every spec, price and model name verified; price dated.
    - B-roll matches the model/colour talked about; product colour accurate on a contact sheet.
    - Pros and cons both present; score matches what was said.
    - `qc.py` passes; tell the user structure, chapters and files.
    
    ## Example requests
    
    - "Edit my phone review - tease the verdict first, spec cards, pros and cons, score at the end."
    - "Sponsored headphone review, add the disclosure and macro B-roll over the talking."
    - "Unboxing + review into a 10-minute video and a 45-second verdict Short."
    

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.