# Brand Launch Film Production > Provider-independent production workflow for AI agents creating brand launch films, product reveal videos, manifesto films, campaign hero films, website hero videos, investor/customer launch assets, and social cutdowns. Use when planning, scripting, generating, editing, reviewing, or delivering launch films that must align strategy, claims, brand voice, visual direction, product proof, accessibil… - **Type:** Skill - **Install:** `agentstack add skill-calesthio-generative-media-skills-brand-launch-film-production` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [calesthio](https://agentstack.voostack.com/s/calesthio) - **Installs:** 0 - **Category:** [Developer Tools](https://agentstack.voostack.com/c/developer-tools) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [calesthio](https://github.com/calesthio) - **Source:** https://github.com/calesthio/generative-media-skills/tree/main/skills/production/content-formats/brand-launch-film-production - **Website:** https://github.com/calesthio/OpenMontage ## Install ```sh agentstack add skill-calesthio-generative-media-skills-brand-launch-film-production ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Brand launch film production Use this skill to make a launch film behave like a business-critical campaign asset, not just a pretty generated video. A strong brand launch film turns positioning into a memorable audiovisual proof: the audience understands who it is for, what changed, why to believe it, and what to do next. This is not legal advice. Treat legal, regulatory, medical, financial, securities, privacy, endorsement, talent, music, likeness, trademark, and platform-policy decisions as approval gates for the client, counsel, or the responsible compliance owner. ## Core production contract Produce a launch film only after you can answer: 1. Who must change their mind or behavior? 2. What is the single launch promise? 3. What proof supports that promise? 4. What must not be claimed, implied, shown, copied, or simulated? 5. Where will the film run, and what cutdowns must ship with it? 6. Who approves strategy, claims, brand, legal/compliance, and final delivery? Separate three categories throughout the work: - **Documented facts:** product features, release date, prices, customer names, results, certifications, patents, case-study numbers, awards, partner logos, platform specs, rights records. - **Approved claims:** claims the client has substantiated and approved for the target market and channel. - **Production heuristics:** creative choices such as pacing, story route, camera style, edit rhythm, music mood, and cutdown structure. Never let a heuristic become an unsupported claim. ## Intake: ask for the launch brief before concepting Collect or infer a concise brief. If critical information is missing, ask only for what blocks a safe first concept pass. Minimum intake: - Product, service, company, or campaign being launched. - Launch type: new company, rebrand, product reveal, feature launch, category entry, fundraising/customer proof, website hero, paid social, event/opening film. - Target audience and decision context: buyer, user, investor, candidate, partner, press, internal team, existing customers, skeptical switchers. - Audience state now: unaware, confused, comparing alternatives, distrustful, price-sensitive, excited but not convinced. - Desired action: sign up, book demo, buy, invest, apply, share, attend, explore website, remember the brand. - Positioning: category, alternative, differentiated promise, emotional territory, forbidden comparisons. - Messaging hierarchy: must-say, should-say, nice-to-say, never-say. - Proof inventory: product footage, screenshots, prototypes, benchmarks, customer quotes, founder story, data, demos, awards, press, certifications, testimonials. - Brand system: strategy, voice, logo rules, colors, typography, motion language, photographic style, sonic identity, accessibility requirements. - Deliverables: master duration, aspect ratios, cutdowns, subtitles, captions, thumbnails, stills, textless versions, clean audio stems, platform placements. - Constraints: deadline, budget, generation providers, source assets, legal review, regulated categories, geographic markets, languages, music/talent rights. If the user asks for immediate generation without enough proof, make a low-risk concept/treatment only. Do not fabricate product capabilities, customer outcomes, endorsements, or certifications. ## Strategy spine Write the strategy spine before scripting: ```text Audience: They currently believe: We need them to believe: Single launch promise: Proof ladder: 1. Product proof: 2. Human proof: 3. Market/category proof: Desired feeling: CTA: Claims requiring approval: Things to avoid: ``` The best launch films usually have one dominant promise and two or three proof beats. If the brief has more than one audience, create audience-specific cutdowns rather than forcing all audiences into one master. ## Choose the story route Present 2-4 concept routes when the creative direction is open. Name each route, give a one-sentence premise, explain why it fits the audience, list proof required, and flag compliance or production risks. Useful routes: - **Manifesto:** belief-led film for a rebrand, category creation, or movement. Best when the brand has a sharp point of view. Risk: vague grandiosity without product proof. - **Product reveal:** mystery-to-reveal structure around the product or service. Best when the object/UI/experience is visually distinctive. Risk: overpromising before proof exists. - **Problem/relief:** opens with a felt customer pain and resolves through the launch. Best for conversion. Risk: manipulative or fear-based framing. - **Founder conviction:** uses a founder or team story to create trust. Best for early-stage, investor, recruiting, or premium craft brands. Risk: self-importance; needs audience relevance. - **Customer transformation:** centers a customer/user before-and-after. Best with authorized testimonials or case studies. Risk: endorsement disclosures and substantiation. - **Demo-as-drama:** turns product interaction into a cinematic sequence. Best for software, devices, tools, and workflows. Risk: fake UI, impossible latency, misleading results. - **Category contrast:** compares old way vs new way without naming competitors unless approved. Best for challenger brands. Risk: comparative claims require support. - **Launch event opener:** designed to energize a room, then work online. Best for keynotes and sales kickoffs. Risk: too much atmosphere, not enough clarity. ## Messaging hierarchy and script discipline Build the script as a hierarchy, not a paragraph. 1. **Hook:** create tension, curiosity, or recognition in the first seconds. 2. **Promise:** state the launch idea plainly. 3. **Proof:** show what makes it true. 4. **Meaning:** connect proof to the audience's life or decision. 5. **CTA or memory lock:** close with action, URL, event date, product name, or campaign line. For a 15-second cut, keep one idea and one proof beat. For 30 seconds, allow one setup, one turn, and one proof cluster. For 60-90 seconds, use three movements: world/problem, launch/proof, invitation. Longer manifesto or investor films need clear chaptering and should earn every abstraction with a concrete image or data point. Write voiceover and title cards so either can carry the film if the other is muted. Social and website hero placements often autoplay silently; assume the first pass may be watched without sound. ## Claims and compliance discipline Documented facts from official sources: - FTC advertising substantiation policy says advertisers and agencies need a reasonable basis for advertising claims before dissemination. Verified 2026-07-10: https://www.ftc.gov/legal-library/browse/ftc-policy-statement-regarding-advertising-substantiation - FTC digital disclosure guidance emphasizes clear and conspicuous disclosures; if a required disclosure cannot be made effectively on a device or platform, that platform may not be appropriate for the ad. Verified 2026-07-10: https://www.ftc.gov/business-guidance/resources/com-disclosures-how-make-effective-disclosures-digital-advertising - FTC endorsement guidance requires disclosure of unexpected material connections and warns that advertisers/intermediaries may be liable for deceptive endorsements or missing disclosures. Verified 2026-07-10: https://www.ftc.gov/business-guidance/resources/ftcs-endorsement-guides-what-people-are-asking and https://www.ecfr.gov/current/title-16/chapter-I/subchapter-B/part-255 - FTC guidance and enforcement around AI claims treats deceptive AI marketing as ordinary deceptive marketing; there is no safe harbor merely because AI is involved. Verified 2026-07-10: https://www.ftc.gov/news-events/news/press-releases/2024/09/ftc-announces-crackdown-deceptive-ai-claims-schemes Production rules: - Mark every express and implied claim in the script, visuals, UI, supers, testimonials, captions, and CTA. - Ask for substantiation for superiority, performance, health, financial, environmental, safety, security, AI capability, customer-result, and savings claims. - Do not transform a customer quote into a broader result claim. If one customer saved 20 hours, do not imply every customer will. - Avoid fake dashboard numbers, fake app screens, fake testimonials, fake logos, invented awards, invented press quotes, and synthetic "customer" endorsements unless the client explicitly approves the fictionalization and disclosure. - Keep qualifications close to the claim they qualify, readable, audible where needed, and present in each cutdown where the claim appears. - Do not bury material disclosures in end slates, tiny supers, or captions only when the claim is prominent in voiceover or image. - If an influencer, customer, employee, investor, advisor, expert, or partner appears to endorse the launch, require approval for disclosure language and placement. - If using "first," "only," "best," "fastest," "most accurate," "clinically proven," "guaranteed," "carbon neutral," "secure," "private," or similar claims, trigger substantiation and counsel/client review. Escalate before production when: - The product is in a regulated area: health, finance, insurance, employment, housing, education, children, alcohol, political, environmental, legal, safety, security, or investment. - The film targets or depicts minors. - The film uses real customers, employees, public figures, synthetic likenesses, voice clones, private data, medical/financial outcomes, or third-party marks. - The launch is tied to fundraising, securities, acquisition, public-company communications, or forward-looking statements. - The concept relies on competitor comparison, benchmark data, survey results, certifications, awards, reviews, or scientific claims. ## Brand voice and concept translation Translate brand strategy into usable direction: ```text Voice: Vocabulary to use: Vocabulary to avoid: Sentence length: Humor/irony level: Confidence level: Technical depth: Emotional temperature: Visual system: Color behavior: Typography/title behavior: Composition: Camera/motion: Product/UI treatment: People: Texture/materials: Things this brand would never do: Sonic system: Music role: VO role: SFX role: Silence: ``` Protect distinctiveness. After each route, ask: could this be any other company in the category? If yes, add a product-specific behavior, brand-specific line, proprietary proof, or sharper world view. ## Visual system and product demonstration Choose visuals according to proof needs: - Use real product footage or UI capture when accuracy matters more than spectacle. - Use generated metaphor, cinematic scenes, or motion graphics when the launch idea is abstract or unavailable to film. - Use founder/customer footage when trust and specificity matter. - Use animation or hybrid UI when the product is not yet camera-ready, but label prototype states clearly if needed. For product/service demos: - Script the user journey before generating shots. - Confirm every UI state, result, metric, latency, notification, and integration is real or clearly illustrative. - Keep screen text legible in every target aspect ratio. - Do not show impossible workflows, unavailable features, unapproved partner logos, or misleading before/after comparisons. - Capture "proof moments": click, solve, compare, export, receive result, handoff, outcome. For generated media prompts: - Convert strategy into shot facts: subject, environment, action, camera, light, motion, brand palette, product constraints, forbidden elements. - Use product reference images, logo files, UI captures, packaging photos, or style boards when available. - Keep claims out of image/video prompts unless the generated shot must include approved text. Claim language belongs in controlled script/title layers. - Ask for separate clean plates when title cards, supers, or localized copy may change. - Avoid asking a model to render exact logos, app screens, small UI text, legal disclaimers, or dense typography unless the provider is known to handle text reliably. Prefer compositing text in edit. - Preserve provenance: record provider, model, prompt, seed/job ID when available, source assets, license terms, edits, approvals, and whether AI was used. ## Shot list and asset plan Create a production table before asset generation: | Beat | Purpose | Visual | Source/generation plan | Audio/text | Claims | Approval risk | | --- | --- | --- | --- | --- | --- | --- | | Hook | Stop scroll | Customer struggling with old workflow | Generate metaphor or film source | Title: "Launch line" | Implied pain claim | Low/medium/high | | Proof | Show product | Real UI walkthrough | Screen capture | VO + callout | Feature and time claim | Requires substantiation | | Close | Convert | Product hero + URL | Composite | CTA | Availability/pricing | Requires launch-date approval | Asset plan checklist: - Brand assets: logos, color values, fonts, icon rules, motion rules. - Product assets: UI captures, device renders, packaging, prototypes, screenshots, API diagrams. - People assets: releases, approved names/titles, wardrobe guidance, likeness/voice permissions. - Customer assets: testimonial approval, disclosure language, case-study substantiation. - Music/SFX/VO: licenses, usage territories, paid-ad rights, term length, exclusivity, union/talent constraints if applicable. - Generated assets: prompts, source references, model/provider metadata, content credentials/provenance notes, edit history. ## Audio, music, VO, SFX, and title cards Use sound to clarify the launch, not to compensate for weak messaging. - Music: define energy curve, reference mood, instrumentation, brand fit, edit points, and rights. Avoid famous-song soundalikes unless explicitly licensed and reviewed. - VO: specify persona, pace, accent/language, pronunciation, disclosure needs, and whether the voice should feel founder-led, narrator-led, customer-led, or purely typographic. - SFX: use for product proof, transitions, tactile interaction, and memory cues. Avoid UI sounds that imply unavailable features. - Title cards: make the promise, proof, and CTA readable without sound. Keep legal qualifiers close to qualified claims. - Silence: use pauses for premium launches, reveals, or founder/customer authenticity. ## Accessibility and inclusive delivery Documented facts from official sources: - WCAG 2.2 Success Criterion 1.2.2 requires captions for prerecorded audio content in synchronized media, with limited exceptions. Verified 2026-07-10: https://www.w3.org/TR/WCAG22/ - W3C media accessibility guidance covers captions, transcripts, audio description, and planning. Verified 2026-07-10: https://www.w3.org/WAI/media/av/ - FCC caption quality guidance for covered television programming emphasizes captions that are accurate, synchronous, complete, and properly placed. Verified 2026-07-10: https://www.fcc.gov/consumers/guides/closed-captioning-television Production rules: - Deliver captions or subtitles for every spoken/audio-led version unless the client explicitly declines and the channel permits it. - Include non-speech audio that matters to meaning: music cues, laughter, alarms, product sounds, applause, silence if intentional. - Keep captions away from platform UI/safe-zone overlays and important product visuals. - Use readable contrast and type size across 16:9, 9:16, 1:1, and any website hero crop. - Provide transcript or descriptive transcript for website and investor/customer-facing assets when practical. - Consider audio description or a descriptive transcript when visual-only information is necessary to understand the launch. - Avoid relying only on color, tiny UI text, fast flashes, or dense kinetic type to convey essential information. ## Delivery, platform specs, and volatile facts Platform requirements change. Re-check specs at production time and record the date checked. As of 2026-07-10: - Google Ads/YouTube video ad specs list common aspect ratios suc … ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [calesthio](https://github.com/calesthio) - **Source:** [calesthio/generative-media-skills](https://github.com/calesthio/generative-media-skills) - **License:** MIT - **Homepage:** https://github.com/calesthio/OpenMontage Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-calesthio-generative-media-skills-brand-launch-film-production - Seller: https://agentstack.voostack.com/s/calesthio - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.