Hire an AI video ad script and prompt writer

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A batch of about 3 or 4 scripts, each a different type, with a source on every customer line
• A line-check score for each script, before and after the fixes
• Timed clip prompts and an edit list for each script that passed
• A pre-launch result for every ad, and the reasons for any held out
• How to get the clips made, and what it costs

How it does the job well

• Write for one named audience and one action. Each asset has a single job.
• Use the customer's own words from reviews, calls or surveys before inventing language.
• Every claim needs proof the business can show. Leave unproven claims out or mark them.
• Judge results against the business goal, such as profit or qualified leads, not clicks alone.

Mistakes it avoids

• Writing the same script type five times and calling it a batch.
• Treating a hook rubric from a handful of videos as a rule.
• Making up a customer quote or a founder's words to fill a gap.
• Fixing a script round after round. Three rounds, then hold it out.
• Running this when the offer is about to change, the product is out of stock, or there is no budget to test several videos. Wait.
• Generic copy that could fit any brand.

What a good result looks like

An example, not your task: You run video ads and want a new batch of scripts each week.

• Path taken. Whether the batch used a winners rubric or not, and why.
• The batch. Three or four scripts of different types, each with its opening, sourced customer lines, line-check scores and clip prompts.
• Held out. Each script that failed a check, with the reasons and what would fix it.
• Types left out. Each script type dropped and what it lacked.
• How the clips get made. The tools you can use, what they cost and the first step.
• Files and missing. Each step's saved file and a list of every fact that was missing.

How the work is checked

• Every script is a different type from the others in the batch.
• Every customer line has a source, and no founder or creator line was invented.
• Every script in the batch scored 85 or more with no Truth problem, or is listed under held out.
• Every ad in the batch has a pre-launch result with no stop-ship item.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Claude Hopkins and Eugene Schwartz would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Read the memory. Opens the opening rubric, the script types from earlier batches and the never-again list.
• Learn from the winners. With six or more past videos that have results and words, it builds a rubric of what your best openings do. With fewer, it skips this and says so.
• Pick the script types. Chooses three or four types that differ, from customer quotes, a creator on camera, a two-person talk and a founder.
• Write each script. Uses the writer for each type. A founder script is only written from the founder's own words.
• Check and loop. Scores each script line by line. Under 85, or any wrong fact, and it fixes and checks again, three rounds at most.
• Make the clip prompts. Times each passing script into clips with one prompt per clip and an edit list.
• Check before launch. Runs the pre-launch check. A stop-ship fault gets two fix rounds, then the ad is held out.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id weekly-video-ad-batch-maker only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/weekly-video-ad-batch-maker/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• An export of your video ads with dates, spend and your result
• The product and offer, with what you may claim
• Customer reviews or call notes, with where each came from
• Transcripts of your past videos, if you have them (optional)
• Product photos for the clip prompts (advised)
• Your founder's own words, if you want a founder video (optional)

Helpful context (optional)

• Transcripts of your best and worst videos
• A note of what changed in your offer this week

Tools and access

• Claude Code: It can keep the video-batch folder, read last week's files and run on a schedule.
• Higgsfield: Paste a clip prompt in and pick Seedance, Kling or Veo. It runs on paid credits.
• Meta Ads Manager: Export your video ads with dates, spend and results from here.
• Any good chat AI: Any current chat AI can sort pasted ads, find patterns and grade drafts. Paste transcripts and results as text.
• Meta Ad Library: Shows a rival's active ads and how long they have run, which hints at which ones win.
• Any good chat AI: Claude, ChatGPT and Gemini all follow this skill well, so use the one you already have.
• Meta Ads Manager: Its A/B test splits one audience between your ads and changes one thing, so the result is fair.
• Meta Ads Manager: Its A/B test splits one audience between your openings and changes one thing, so the result is fair.
• Claude: Keeps the speaker rules and the one-detail limit in view across a long brief.
• Veo: Turns a start frame of your AI presenter into short talking clips with the lips matched to the line. Veo is a paid Google service that needs an account; this script does not need it.
• Claude: Keeps every spoken fact tied to your list and counts the words in each clip.
• Kling: Makes clips up to 15 seconds with several shots and lines for more than one speaker, so a clip can hold a few turns.
• Veo: Speaks quoted lines for a named speaker, so each turn comes from the right mouth.
• Claude: Keeps each line traced to the founder's own words and turns gaps into questions instead of filling them.
• Descript: Turns the founder's recorded answers into text you can paste in, then lets you cut the film by editing that text.
• Veo: Follows timed shot lines inside one 8-second prompt and can aim at an end frame you give it.
• Kling: Takes several shots in one generation from a shot list, keeping the same person across the cuts.
• Any good chat AI: Writing the clip plan and shot tables to these rules works well in any strong chat AI.
• Claude: Reads full-size images closely, so it can spell-check the words in the picture and spot the safe zones.
• Meta Ads Manager: Its safe zone guardrail marks the area the app's buttons and labels won't cover, so you see what might get hidden.

Functions and deliverables

• Video ad script and prompt writer: Use customer evidence to draft varied video ads. Get scripts, shot lists and clip prompts with failed drafts held back.
  • A batch of about 3 or 4 scripts, each a different type, with a source on every customer line
  • A line-check score for each script, before and after the fixes
  • Timed clip prompts and an edit list for each script that passed
  • A pre-launch result for every ad, and the reasons for any held out
  • How to get the clips made, and what it costs
• Video hook review checklist writer: Build a winning hook rubric from your top ads and one rival's. Get hook formats, plus your draft hooks graded A to D with three rewrites.
  • Winners and losers split at the break point for each source
  • A rubric in four buckets: psychology, trigger words and framing, grammar, and what is not obvious
  • Rules shared across sources, plus custom rules per source, and fill-in-the-blank hook formats
  • An A to D grade with reasons, and three rewrites, for each draft hook
• Video ad format planner: Share your buyer and proof. Get up to four supported video formats, with an opening, tone and reason for each.
  • A batch of script types, each with its group, selling beat, opening and tone
  • The fact that meets each type's needs, with its source
  • A list of skipped types, each with the reason
  • The scripts themselves, if you pair it with a writing skill
  • Any of its own checks the batch couldn't pass, named one per line
• Customer quote video ad writer: Write a video ad script from real customer quotes: nine beats with a source for every line, checked openings and a first test plan.
  • The belief shift and a list of what was assumed
  • The script, beat by beat, with spoken line, on-screen text, picture and source
  • The spoken word count against the budget for the length and pace
  • 4 to 6 openings from different sources, each marked pass or fail on four checks
  • A suggested format, a test plan and a list of gaps
  • Any of its own checks the script still misses, one line each
• Selfie-style video ad script writer: Write a selfie-style video ad script from your buyer's doubt and product facts. Get five timed beats, a shot list and two other openings.
  • The story in one line, then a five-beat script with what's said, on-screen text and a picture note per beat
  • The one detail that carries the ad, the length worked out from the steps, and a word count that fits it
  • A shot list saying which hand holds what, delivery notes and two other openings
  • A cast note on who should carry the ad and where the script goes next, plus every claim with its source
  • Any of its own checks the script still misses, one line each
• Podcast-style video ad script writer: Write a podcast-style video ad script for two made-up speakers. Give it a buyer doubt and your facts. You get a timed script with sources.
  • A belief line and the way the argument moves
  • A line-by-line script for Host and Guest, with picture notes
  • A clip plan with the words in each clip, a muted opener and the end text
  • Every fact said, with its source, the disclosure line and a cast note
  • Any of its own checks the script still missed, one line each, or none
• Founder video ad script writer: Write a founder video ad script built on one real decision. Give it your founder's words and product facts. You get a filming brief.
  • A filming brief with each line's move, words and what to film
  • Two or three openings to record, each true for your founder
  • Interview questions, if your founder's words are missing, and notes on the proof behind each claim
  • A casting and wardrobe note on who appears, where they film and what they wear
• Video ad script line checker: Check your video ad script line by line. Get each weak line with a fix from your own facts, plus a score out of 100 before and after.
  • Every weak line, quoted, with the check it fails, what's wrong and a fix
  • The beats the script is missing
  • The spoken word count against the budget for the length
  • A list of facts a fix would need that you haven't given
  • A score out of 100 now and after the fixes, with the sum shown
• Video ad shot list maker: Share an approved script and visual references. Get timed shot tables, frame prompts, clip prompts and an edit list for production.
  • A clip plan and a film plan for each clip
  • A shot table per clip with timings, size, camera position, move and one action, for your approval
  • Start and end frame prompts and clip prompts that describe only the motion
  • A voice plan and an edit list with captions, music and the end card
  • Any of its own checks the shot list couldn't pass, named one per line
• Pre-launch ad checker: Share your ad, audience and approved facts. Get located issues, fixes, human checks and a custom score before launch.
  • A version-bound list of observed problems and the smallest fixes
  • The buyer argument checked against the supplied facts and proof
  • A clear list of media, destination, tracking or spec checks still needed
  • A scoped verdict and observed/eligible coverage
  • An internal editing score, with any post-fix score marked provisional
• Brand profile builder: Share your website. Get a reusable profile covering your products, customers, voice and approved claims, with sources.
  • A folder of eight plain files, from a start file for your AI to a claims list and a change log
  • A tag on every fact, and every claim marked approved, pending or banned
  • A two-sentence diagnosis of your biggest gap, and a list of what's missing and who can answer it
  • A short slice for your next step, if you name one
• Claude Hopkins: Write specific ads or plan a small buyer test using Claude Hopkins’s recorded approach. Replace vague praise with facts and track cost per customer.
  • Your problem worked on the way Claude Hopkins would: their questions, their method, their push-back
  • The answer or the work, in Claude Hopkins's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Eugene Schwartz: Match a headline or sales page to what buyers already want and know. Apply Eugene Schwartz’s recorded approach to research, claims and proof.
  • Your problem worked on the way Eugene Schwartz would: their questions, their method, their push-back
  • The answer or the work, in Eugene Schwartz's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• You run video ads and want a new batch of scripts each week.
• You have customer reviews but your scripts all sound the same.
• You want every line checked against your facts before you pay for clips.

Limits to check before starting

• You want still image ads: this agent only plans video, so use the Image ad remake planner.
• You need one script type only, like a founder video: write it with the Founder video ad script writer.
• Your offer is about to change, the product is out of stock, or you have no budget to test: wait, or plan with a person first.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Start a chat with the agent. In Claude Code, add it as a skill or save it in .claude/agents, then ask it to make this week's batch.
• Attach last week's export, your customer words and the product name. Add transcripts and photos if you have them.
• Read the batch. Say yes or no to each script.
• If you want clips made, pick a tool. The agent asks before it uses any credits.
• Run it again next Monday. It reads its memory folder to see which types did well.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/weekly-video-ad-batch-maker for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.