/04-content-brief

description: For one priority from content_priorities.json, generate a content brief with explicit slots for the human expert (founder/domain specialist) to fill with real cases, real data, and real opinions. This is the human-in-loop checkpoint — the moat that prevents the entire workflow from collapsing into "AI writing more AI-readable AI slop." Outputs briefs/<priority-id>.md. Use after 03-gap; consumed by 05-production ONLY after the expert has filled the marked slots.
argument-hint: "<client-folder, e.g. clients/acme>"

04 · Content Brief — Human-in-Loop Checkpoint

This is the most important command in the pipeline. Without it, the system collapses into the same generic AI-content-mill that has saturated the GEO commodity layer. With it, the system produces content AI engines actually cite — because it contains things AI can't auto-generate.

**The principle**: the pipeline builds the scaffold; humans fill the soul. This command outputs scaffold with explicit, named, schema-tight blank spaces. Until those blanks are filled, 05-production refuses to run.

---

Inputs

• `clients/<slug>/brand_context.json`
• `clients/<slug>/content_priorities.json`
• A specific `priority_id` (passed by user or auto-picked: highest unbriefed).

Output

• `clients/<slug>/briefs/<priority-id>.md` — human-readable brief with

REQUIRED-FILL slots clearly marked.

• `clients/<slug>/briefs/<priority-id>.meta.json` — machine-readable

metadata: which slots are filled, by whom, when.

• `clients/<slug>/briefs/<priority-id>.html` — interactive fill-form for

the client/expert (rendered by the `geo-review-html` skill; this is the artifact the reviewer actually opens, not the `.md`).

---

Procedure

Step 1 — Load context

jq --arg id "<priority-id>" '.priorities[] | select(.id==$id)' \
  clients/<slug>/content_priorities.json

Pull the priority record. Cross-reference brand_context for the relevant ICP, voice samples, and competitor positioning.

Step 2 — Choose the angle

Generate 2–3 candidate angles for the priority. An angle is a one-sentence statement of the unique thesis the piece will defend. Bad angle: "Best CRM for small business." Good angle: "Why bookkeepers (not sales teams) should own CRM choice in firms under 20 people, based on our 200-firm dataset."

The angle MUST satisfy:

• Aligns with `core_differentiator` and `competitive_moat` from brand_context.
• Targets the priority's `intent` directly.
• Cannot be written without the brand's specific

expertise/data/perspective (i.e., GPT-4 with a Google search couldn't produce it).

Present the candidate angles to the user. Wait for selection or for the user to propose a fourth angle.

Step 3 — Build the brief skeleton

Use the `recommended_format` from the priority to pick a template:

• `comparison-page` → table-driven, ≥3 alternatives, decision rubric.
• `deep-guide` → outline with H2/H3, FAQ block, glossary.
• `case-study` → setup → action → result → lessons.
• `data-report` → headline finding → methodology → data tables → caveats.
• `expert-essay` → POV statement → evidence → counter-arguments → conclusion.
• `definition-page` → canonical definition → variants → examples → related concepts.
• `how-to` → prerequisites → steps → verification → troubleshooting.
• `list-roundup` → criteria → ranked list → notes per item.
• `faq-block` → 8–15 Q&A pairs.

Step 4 — Insert REQUIRED-FILL slots

For each `required_assets` entry in the priority record, insert a clearly-marked slot in the brief. Each slot has:

> **REQUIRED-FILL · {slot_id} · {asset_type}**
>
> **What goes here**: {1-2 sentence description of what the expert
> must provide. Be specific.}
>
> **Why it matters for GEO**: {1 sentence — what this real input does
> for citation-worthiness, e.g., "AI engines cite primary data 3.2× more
> than rephrased data." (cite Princeton KDD 2024 finding.)}
>
> **Bad fill (rejected)**: {example of what NOT to write — usually a
> generic placeholder or AI-paraphrased text.}
>
> **Good fill (accepted)**: {example showing the level of specificity
> required.}
>
> **(Expert: replace this entire block with your content. Do not delete
> the slot_id — 05-production reads it to verify the slot is filled.)**

Slot ids: `data-1`, `quote-1`, `case-1`, etc. — stable, sequential per type.

Step 5 — Mandatory GEO elements

Every brief must include scaffolding for these (Princeton KDD 2024 shows they boost AI citation rate):

1. **Statistics block** — at least 2 quantitative claims (REQUIRED-FILL slots if not already in brand_context). 2. **Quotation block** — at least 1 direct quote from a named expert (REQUIRED-FILL slot). 3. **Citation block** — list of 5–10 external authoritative sources the piece will cite. Pre-populate from brand_context Layer 2 `citation_sources` where relevant; mark new ones as REQUIRED-FILL. 4. **FAQ block** — 5–8 Q&A pairs derived from `real_user_questions`.

Step 6 — Voice-match notes

Pull `voice_samples` from brand_context. Add a "Voice notes" section at the top of the brief instructing the writer (05-production or human) to match: tone attributes, sentence length range, jargon level, first-person vs third-person.

Step 7 — Write the brief and meta

Brief MD layout:

# Brief — <priority.query>

> Priority id: <id>
> Generated: <ISO ts>
> Format: <recommended_format>
> Status: awaiting-expert-fill

## Angle
<chosen angle, one paragraph>

## Voice notes
<bullet list>

## Outline
<H2/H3 structure with REQUIRED-FILL slots inline>

## Mandatory GEO elements
### Statistics
<slots>
### Quotations
<slots>
### Citations
<pre-populated + slots>
### FAQ
<slots derived from real_user_questions>

## Distribution targets
<from priority.recommended_format + intent — pre-suggested>

## Sign-off checklist (for expert)
- [ ] All REQUIRED-FILL slots replaced
- [ ] Angle still feels right after filling
- [ ] Statistics have sources or methodology notes
- [ ] Quote