Faceless & Automation Channels prompts

Help me choose a niche for a faceless YouTube channel.

NICHES I AM CONSIDERING: [LIST 3-6 NICHES]
MY SKILLS AND INTERESTS: [WHAT I KNOW OR ENJOY]
BUDGET PER MONTH FOR TOOLS/VOICE/FOOTAGE: [AMOUNT]
HOURS PER WEEK: [HOURS]
LANGUAGE AND AUDIENCE REGION: [e.g. English, US/UK]

Score each niche 1-5 on:
1. DEMAND - are small channels in this niche getting views? Tell me what to search on YouTube to check.
2. ORIGINALITY ROOM - can I add real value (research, analysis, a unique format), or does it rely on reused clips and templated scripts? This matters because YouTube's monetisation policies do not pay for mass-produced, repetitive or reused content.
3. PRODUCTION COST - footage, animation, research time per video
4. ADVERTISER FRIENDLINESS - topics like finance and tech usually pay better than entertainment; topics like violence, tragedy or shock content can be limited
5. MY EDGE - how my skills help

Then:
- A comparison table with total scores
- Your recommended niche with a one-paragraph reason
- 3 formats that would make my channel different from template channels in that niche
- The top 3 risks, including policy risks, and how to avoid them
- What the first 10 video titles could be

Be blunt. If a niche is saturated with near-identical channels, say so.

Write a script for a faceless YouTube explainer.

TOPIC: [TOPIC]
ANGLE (what makes this version different): [ANGLE]
LENGTH: [MINUTES]
VOICE: [AI voice / my own voice], tone: [calm narrator / documentary / energetic]
SOURCES I HAVE: [PASTE LINKS, NOTES OR FACTS]

Write the script as a table with these columns:
Time | Narration | Visual (what is on screen) | Source (for any factual claim)

Rules for narration:
- Hook in the first 10 seconds with the most surprising true fact or question
- Short sentences that sound natural when read by a voice, no tongue-twisters, spell out numbers and abbreviations the way they should be spoken
- A re-hook every 60-90 seconds
- One clear idea per scene

Rules for visuals:
- Suggest a mix: stock footage, simple motion graphics, maps, charts, text on screen, and original AI-generated images
- Avoid anything that needs copyrighted clips (movies, TV, other creators)
- Write each visual so it can be searched on a stock site or written as an image prompt

Rules for facts:
- Only use facts from my sources. Where a fact would help but I did not provide a source, write [NEEDS SOURCE: what to look up].
- Never invent statistics, dates or quotes.

After the table, give me: a list of claims to double-check, 5 title options, and a one-sentence description of what makes this video original (useful if YouTube ever reviews the channel).

Design a production pipeline for my faceless YouTube channel.

NICHE AND FORMAT: [NICHE, VIDEO STYLE, LENGTH]
TARGET OUTPUT: [VIDEOS PER WEEK]
TEAM: [just me / me + freelancers - list roles]
TOOLS I ALREADY USE: [LIST]
BUDGET: [MONTHLY]

Produce:
1. THE PIPELINE - numbered stages: idea > research > script > fact-check > voiceover > visuals > edit > thumbnail > metadata > upload > review. For each stage: input, output, who does it, tool type, and time estimate.
2. QUALITY GATES - the checks that must pass before moving on (e.g. every claim has a source, audio has no mispronunciations, no copyrighted footage).
3. TEMPLATES I NEED - list documents to create once and reuse (script template, thumbnail template, description template, checklist).
4. WHERE AI HELPS AND WHERE IT SHOULD NOT - be specific. AI can draft, summarise, suggest visuals and write metadata. Human review should own facts, final script voice, and anything that makes the channel original.
5. A WEEKLY SCHEDULE that hits my output target within my hours
6. THE BOTTLENECK - which stage will slow me down first, and how to fix it

Keep tool suggestions generic (e.g. "AI voice generator") unless I named a tool, and remind me to check each tool's commercial-use licence.

Rewrite this script so it sounds natural as a voiceover.

SCRIPT: [PASTE SCRIPT]
VOICE: [AI voice name/type or human narrator], pace: [slow / medium / fast]
AUDIENCE: [WHO]

Rewrite rules:
- Sentences under 20 words. Break long ones.
- Write numbers, units and abbreviations as they should be spoken ("twenty twenty-six", "per cent", "U.S." -> "US" or "the United States").
- Add pauses using line breaks or "..." where a narrator would breathe.
- Mark emphasis with *asterisks* on the one word per sentence that carries the meaning (remove them if the voice tool does not support emphasis).
- Remove written-only phrases: "as mentioned above", "the following", "see below".
- Remove filler openers: "In this video", "So,", "Basically".
- Keep all facts and meaning exactly the same. Do not add new claims.

Then give me:
- A PRONUNCIATION LIST - names, places or technical words the voice might get wrong, with phonetic spellings
- The estimated read time at my chosen pace
- 3 places where adding background music or a sound effect would help, and 1 place where silence is better

Create a visual plan for my faceless video.

SCRIPT (with timestamps if possible): [PASTE]
VISUAL STYLE: [e.g. cinematic stock footage / flat illustration / 3D renders / documentary archive look]
TOOLS: [e.g. stock library, image generator, video generator]
ASPECT RATIO: [16:9 / 9:16]

For each scene or every 5-8 seconds of narration, give:
1. TIME
2. NARRATION LINE (short)
3. SHOT TYPE - wide, close-up, overhead, text card, chart, map
4. STOCK SEARCH TERMS - 2-3 specific phrases to search on a stock site
5. AI IMAGE OR VIDEO PROMPT - one prompt describing subject, setting, lighting, camera and style, consistent with my visual style. Keep the same style words in every prompt so the video looks consistent.
6. MOTION - static / slow zoom / pan / animated

Rules:
- No real people's likeness, brand logos, or copyrighted characters in AI prompts.
- Avoid visuals that literally repeat the words; show what the narration means.
- Every 20-30 seconds, suggest a pattern break: a chart, a map, a text card, or a change of shot type.
- End with a STYLE BLOCK: the exact style words to reuse in every prompt so all visuals match.

Review my faceless video for monetisation and copyright risks before I upload.

SCRIPT: [PASTE]
ASSETS USED: [LIST: stock footage source, music source, AI tools, any clips from others]
TITLE AND THUMBNAIL IDEA: [DESCRIBE]
HOW THIS VIDEO IS DIFFERENT FROM OTHER VIDEOS ON MY CHANNEL: [DESCRIBE]

Check and report on:
1. REUSED CONTENT - does it rely on other people's clips, readings of articles, or compilations without significant original commentary?
2. REPETITIVE CONTENT - would this video look near-identical to my other uploads (same script template, same visuals, only names swapped)?
3. ADVERTISER-FRIENDLY ISSUES - profanity, shocking imagery, sensitive events, misleading claims in title/thumbnail
4. COPYRIGHT - music and footage licences, and whether AI tools' terms allow commercial use
5. SYNTHETIC CONTENT DISCLOSURE - whether any realistic AI-generated or altered people, places or events mean I should use YouTube's altered or synthetic content label
6. MISLEADING PACKAGING - does the title or thumbnail promise something the video does not show?

For each: risk level (low / medium / high), the specific line or asset, and the fix.

Finish with a note: this is a checklist, not legal advice, and YouTube's current policies are the final word - tell me which official YouTube policy pages to read for each area.

Design a repeatable video series for my faceless channel.

NICHE: [NICHE]
AUDIENCE: [WHO]
WHAT VIEWERS LIKED MOST SO FAR: [COMMENTS, BEST VIDEOS, or "new channel"]

Design the series:
1. SERIES NAME and one-line promise
2. THE FIXED PARTS - what stays the same every episode (structure, intro length, visual style, sign-off). These build recognition.
3. THE VARIABLE PARTS - what must change every episode (the story, the research, the angle, a new data point, a new example). These keep it original.
4. EPISODE TEMPLATE - timed sections for a typical episode
5. 15 EPISODE IDEAS - each with a unique angle, not just a new name dropped into the same script
6. THE VARIETY TEST - for any two episodes, list what a viewer would notice is different. If the answer is "only the topic name", the series is too templated - fix it.
7. FORMAT UPGRADES for episodes 10+ so the series evolves (a data segment, viewer questions, a comparison, an update to an older episode)