AI Creative Brief System

Both - depending on which tier you buy.

Blueprint ($997) generates the structured briefs - the spec that drives high-performing image and video creative. Audience, pain, hook, angle, visual direction, CTA, technical specs (aspect ratio, channel, format) - every field filled live from real community research. You run the actual creative through Midjourney, Runway, or your in-house tool. The bottleneck for performance creative is brief quality, not pixel generation.

Blueprint + Agent Codebase ($2,997) ships the full production pipeline: seven AI image models, Veo and Kling for video, AI voiceover, compositing, animated captions, and every aspect ratio (9:16 / 16:9 / 1:1). A full 80-image Demand Gen pack costs $4-14 in API fees and lands in under an hour. Video runs end-to-end, from script to YouTube upload.

The agent pulls from public community data - relevant subreddits, public Discord servers, review aggregators (Trustpilot, Google Reviews, Yotpo) - and reads the product page directly for marketing claims and spec data.

What comes back is real customer language, not template phrases. The actual words your buyer uses for the pain you solve. The objections they raise. The transformation language they describe. The before-state vocabulary that signals problem-aware vs solution-aware traffic.

Optional: connect Klaviyo or Gorgias for first-party customer language from your actual buyers. The brand-voc-mine command (Agent Codebase tier) harvests buyer language across reviews, support tickets, and community data into a structured voice-of-customer library that gets reused across every future brief.

The system encodes 60+ proven hook and angle patterns mapped against three axes:

Awareness levels - the five Schwartz stages from unaware through most-aware.

Traffic sources - cold paid, retargeting, organic, email, UGC.

Creative types - static image, motion graphic, talking-head UGC, demo, founder-led.

Concretely: pattern interrupts, problem-amplification opens, before/after splits, "I was X until Y" first-person openers, deductive hooks ("the reason X is..."), curiosity gaps, social-proof leads, contrarian openers - and 50+ more. Each pattern ships with a template, an example, and a list of categories where it tends to win.

The agent picks the right pattern for your audience and awareness level; you can override or pin a specific pattern you want tested.

Every brief and every generated asset is scored against the 13-Point Creative Quality v10.5 framework before it ships. Each point gates the asset; a failed point either fixes automatically or blocks and escalates:

(1) Hook clarity in the first 1.5s · (2) Pattern interrupt · (3) Audience-language fidelity · (4) Pain naming · (5) Mechanism specificity · (6) Proof anchor · (7) Visualizable claim · (8) Falsifiable claim · (9) Visual-to-claim alignment · (10) CTA-to-hook closure · (11) Aspect-ratio-correct composition · (12) On-brand asset usage · (13) Direct-response discipline (no aspirational fluff).

The creative-refresh-review command (Agent Codebase tier) scores every new asset against this rubric pre-ship and blocks anything that fails. Nothing reaches your ad account without passing.

Yes - and it's one of the system's strongest use cases.

The creative-brief-ugc command (Agent Codebase tier) generates UGC creator briefs with hooks, talking points, shot lists, and anti-polish rules. Briefs are formatted for direct creator hand-off - you send the brief to your UGC vendor or in-house creator and they shoot to spec without having to interpret a vague Slack message.

The hook + angle library is the same one used for AI-generated creative, so your UGC and AI creative stay aligned on the same messaging frameworks. Switch a brief from AI to UGC in one flag, keep the same angle.

No - it compresses the research and drafting work so they spend their day on judgment instead of production.

The work that used to take a strategist two days (audience research, angle exploration, hook drafting, brief writing) collapses to 20-30 minutes. Your strategist then spends their day on the harder calls: picking the right angle for this week's test, working with creators, reviewing variants, making the calls the agent can't - brand voice, partnership-specific constraints, things only a human knows about your business.

Most teams we work with run 3-5× more concepts per week with the same headcount after deploying this. The bottleneck moves from production to learning rate.

Built for the major paid social and demand-gen channels:

Meta (Facebook + Instagram) - full image and video pipeline, Reels, Stories, feed.

TikTok - UGC briefs, hook-led video, native-style creative.

YouTube (Demand Gen, In-Stream) - long-form briefs, multi-aspect generation.

Google Demand Gen - image and video pack assembly with channel-control defaults baked in. One example: Shorts stays off by default, which in our published benchmark cut CPA from ~$245 to $94.

Pinterest - image-led briefs and a prompt library tuned for the aesthetic.

The Agent Codebase tier includes demandgen-prep-images, demandgen-prep-videos, demandgen-yt-videos (YouTube upload + campaign config generation), and demandgen-autogrow for closed-loop creative refresh.

Instant access to the 24-module curriculum (Creative Scoring System + Demand Gen Deployment System + Automation Reference), the 9 implementation templates, and 2 bonuses (AI Image Prompt Library, Creative Formats Swipe File). The Agent Codebase tier adds the Git repo with 45 skills and the full theme libraries: 49 image themes, 22 video themes with source clips and video-theme.json definitions, the hooks library, and the audio SFX library.

A 5-email post-purchase sequence walks you through your first 30 days. Lifetime updates ship free, forever. We push 1-2 minor versions per month - the 2026-04 update added Module 23, Module 24, the AI video pipeline, and the demandgen-autogrow loop, all included in your purchase.