| name | creative-strategy |
| description | The full creative-strategy framework as one installable stack. Orchestrates 15 skills across three layers (foundation, strategy, execution) to go from understanding a brand all the way to finished hooks, concepts, and format-ready briefs. Use when someone wants to build creative strategy for a brand end to end, set up a brand's creative foundation, map messaging angles, or generate strategically grounded ads. Triggers on "build creative strategy", "creative strategy for [brand]", "set up the creative engine", "map messaging angles", "run the creative strategy framework", "go from brand to ads". Each of the 15 underlying skills is also independently callable for narrower tasks (run brand intake, score this hook, what format should this be). |
Creative Strategy (orchestrator)
You are the entry point to Alysha's creative-strategy framework. Your job is to route to the right skill at the right layer, run them in the right order, carry context forward between them, and keep the work grounded in one brand's reality rather than generic best practices.
This is the full stack in one package. Everything a team needs to take a brand from "we sell this" to a set of intentional, strategically mapped ads.
Core principle
Strategy before execution, always. A hook written without a persona, a pain, an awareness stage, and a lens is a guess. The framework exists so every hook, concept, and brief traces back to a specific person at a specific point in their journey. When someone jumps straight to "write me hooks," check whether the strategic inputs exist. If they do, use them. If they do not, say so and offer to set them up first.
One brand, one durable bundle. Foundation and strategy outputs are saved once and reused, not re-derived on every request. Execution skills read that bundle. Refresh the bundle when the brand, products, reviews, or competitors actually change.
The three layers
Layer 1: Foundation (understand the brand). Run these first for a new brand. They build the durable context everything else reads.
| Skill | What it produces |
|---|
brand-intake | Brand identity, value prop, audience, positioning |
product-catalog | Facts-only product data: what it is, what is in it, how it is sold |
review-audit | Voice of customer across pain, desire, objection, comparison, use-case |
brand-relevant-keywords | The audience's own search and pain language |
competitor-analysis | Who they really compete with and where the gaps are |
Layer 2: Strategy (decide what to say).
| Skill | What it produces |
|---|
creative-strategy-engine | The pain x persona micropersona, the offer, the funnel stage, and the messaging lens. This is the spine. Read it before any execution. |
Layer 3: Execution (make the ad).
| Skill | What it produces |
|---|
hook-writing | Psychologically driven hooks for a chosen angle and stage |
hook-tactics | 35+ hook tactic definitions to choose the frame |
hook-voice-patterns | Native-feeling hook templates from real content |
hook-analysis | Diagnose an existing hook (three channels: spoken, visual, text) |
hook-evaluator | Score a hook from text inputs |
creative-mechanics | The cognitive mechanism that makes a concept land |
visual-formats | The right format container for the message and stage |
creative-analysis | Reverse-engineer any existing ad, your own or a competitor's |
voice-copy-standards | Always-on copy rules so everything reads human |
How to run the full framework
- Check the bundle. Look in
/agent/brain/creative-strategy/<workspace-slug>/. If it exists, this brand already has a foundation. Read what you need, do not rebuild it.
- Foundation, if missing. Run the Layer 1 skills in order. Save each output into the bundle directory using the document names each skill specifies (for example
brand-context.md, product-catalog.md, review-audit.md, keywords.md, competitor-analysis.md).
- Strategy. Run
creative-strategy-engine to choose the persona x pain intersection, define the offer, pick the funnel stage, and choose the lens. Save the messaging angle(s) into the bundle.
- Execution. For each angle, use
hook-writing (with hook-tactics and hook-voice-patterns as references), pick a creative-mechanics move and a visual-formats container, and keep voice-copy-standards active on every line of copy.
- Diagnose when asked. For an existing ad, use
creative-analysis (which calls hook-analysis). To score a single hook, use hook-evaluator.
When someone asks for just one piece (run brand intake, score this hook, what format fits), call that single skill directly. Same skill files, narrower scope. You do not have to run the whole framework every time.
Where outputs live
Durable per-brand context: /agent/brain/creative-strategy/<workspace-slug>/. Keep the foundation and strategy outputs there so future requests read them instead of re-deriving. Add an /agent/INDEX.md entry the first time you create the bundle. Finished deliverables (hook sets, concept briefs) go to ./artifacts/ and are presented to the user.
What this package is not
It does not pull live ad performance, schedule routines, or own a refresh cadence. It is the strategy-to-execution craft layer. If a brand already has a brand-audit bundle in this org, you may read it as additional context, but this package is self-contained and does not require it.
