Discovery

Discovery is the foundation of the whole workflow and the most important part of setup. The discovery skills build the durable context every other skill reads — and they are what make the difference between output that sounds like you and output that sounds like generic AI marketing copy.

Don’t skip this

The production skills work without configuration, but without purpose, content buckets, and especially voice/tonality, they fall back to safe, generic defaults — drafts drift toward a sales-y, everyone’s-voice register. The discovery skills calibrate the agent to your actual audience, territory, and voice. A few interviews up front pay off across every post you write afterward.

The discovery skills are interviews, not forms. The agent asks questions, reflects a draft back to you, and iterates until the result genuinely sounds like you. Run them in order — each one builds on the last.

1. discover-purpose → purpose.md

Why first: purpose is the filter everything else passes through. Without it, ideas and angles have nothing to align to.

It moves from a broad opening (“why are you thinking about writing online right now?”) through the dimensions that are most alive for you — motivation, audience, category, point of view, style (educating vs. entertaining), and vision — then reflects back a draft you sharpen until it’s yours.

Produces purpose.md: the labeled dimensions, a natural-language Purpose Statement, and a five-question Decision Filter the other skills use to judge whether a topic is worth your time.

Run discover-purpose with me — help me define why I write online and who it's for.

2. discover-buckets → expertise.md + buckets.md

Why second: once you know why you write, you decide what you own. This stops you from writing about everything and being known for nothing.

It lists 10–20 of your expert zones (breadth first), finds your One Big Key Zone, overlaps it with 2–3 secondary zones to locate your genius zone (the unfair-advantage intersection), then translates that into content buckets (General / Niche / Industry) and stress-tests them for sustainability.

Produces expertise.md (the full zone list) and buckets.md (your buckets and topic territory), both read by ideate and the drafting skills.

Run discover-buckets — help me map my expertise into content buckets and find my genius zone.

3. discover-tonality → tonality.md

Why third (and highest-leverage): this is the file that keeps every draft sounding like you. It needs purpose.md and buckets.md so its examples are drawn from your real domain.

It weaves preference questions, comparisons, and rewrite prompts to extract your values across six voice dimensions — Commitment, Reasoning Style, Reader Relationship, Emotional Register, Density, and the failure modes the agent should resist — then checks how your voice shifts by piece type and synthesizes a profile you confirm.

Produces tonality.md: a voice summary, dimension profiles, agent-specific failure modes, register tendencies, format rules, and 8–10 reference samples from your own rewrites. Every content skill loads it to match your voice and check output against your anti-patterns.

Run discover-tonality — build my voice profile from examples so drafts sound like me.

Writing in German?

Run finde-stil instead of discover-tonality. It produces stil.md from scratch in German and is independent of tonality.md.

When discovery is “done”

You have a populated purpose.md, buckets.md, and tonality.md (or stil.md) that you actually recognize as yours. They’re re-runnable — revisit any of them later to refine the foundation as your writing evolves.

With the foundation in place, move on to Production. For each interview’s details, see discover-purpose, discover-buckets, and discover-tonality.