Packaging Content for SEO

Required Background

Look for this project's brand-kit reference material — commonly produced via the building-brand-kits skill — for the facts this skill needs:

• Fact sheet — for the org's current product/service lines (used to derive keyword focus), canonical URLs, and owned channels (used to find internal-link targets).
• Voice card — for which named/bylined contributor or institutional voice applies to the piece, which drives the author-bio block and the register the title/meta should match.
• Media-coverage inventory — if the input is press-sourced, for the clip's logged outlet/date/angle.

If none of this exists, ask the user for the relevant facts or flag the gap explicitly — never invent stats, URLs, contributor names, or a keyword list. This skill applies that material; it does not redefine it.

Overview

A surprisingly common gap in otherwise decent writing is structural, not stylistic: titles that lead with the brand name instead of a keyword, meta descriptions that are just a comma-separated keyword dump, zero real H1/H2/H3 structure in the CMS's section-heading fields, no internal linking, and press hits that never get repurposed into a backlink. This skill packages a finished draft or clip into those pieces — title, meta description, section-heading map, internal links, an optional press-repurposing note, and an author-bio block — without touching the body prose or voice.

When to Use

• A blog post or static page is finished or near-finished and needs a title, meta description, section headings, and internal links before publishing.
• A press-clip URL or text needs a short SEO repurposing note (pull-quote + commentary + backlink) — not a full repurposing package; hand that off to repurposing-press-coverage.
• Not for: rewriting or "improving" the draft's actual voice or content — that's auditing-brand-voice's job. This skill only ever touches title/meta/headings/links.

Quick Reference

On the CMS: the exact field names for title, meta, and section headings vary by platform (Strikingly, Webflow, WordPress, Squarespace, a custom CMS...) — find where this project's CMS exposes those fields, but the content rules in this skill don't change based on which one it is.

On keyword focus: don't assume a fixed keyword list. Derive the current focus from the project's fact sheet's product/service lines, or ask the user which keyword(s) this specific piece is actually about.

Workflow

1. Read the input. Get the full draft (or press-clip text/URL) and identify which page(s) or section(s) it should support.
2. Identify voice. Check the project's voice-card for whether this piece carries a named contributor's voice or an institutional voice — this determines the author-bio block and the register the title/meta should match (e.g. consumer-facing vs. B2B/partner-facing vs. no-byline institutional).
3. Draft the title. Lead with a keyword the draft is actually about — drawn from the fact sheet's product/service lines, or confirmed with the user — not just the org name alone. ≤60 chars.
4. Draft the meta description. Write it as a real sentence a person would read, summarizing what the page delivers, 120–158 chars. Never string keywords together with commas.
5. Map section titles. Go through the draft's actual paragraph breaks and assign one H1 (page title, usually the same as the title) + H2s for each major section + H3s for sub-points, if any. This is exact text to paste into the CMS's section-heading fields — not a suggestion to restructure the draft.
6. Find 2–5 internal links. Match draft content to existing pages and related blog posts. Write natural, descriptive anchor text — never "click here" or a bare URL. For example, if the draft mentions a program the org runs, link it to that program's page with anchor text like "our legacy-storytelling workshops" pointing at example.org/programs/legacy-storytelling, rather than a generic "learn more" pointing at the homepage. Link with the pillar-page / topic-cluster model in mind: point supporting posts at the broad "pillar" page for their topic and back, so related content reinforces one hub rather than scattering authority across disconnected pages.
7. If press-sourced: add a short repurposing note — one pull-quote from the clip, 1–2 sentences of original commentary framing why it matters, and a backlink to the outlet. Check the media-coverage inventory for the clip's logged details (outlet, date, angle) rather than inventing them; if it isn't logged, ask for the source instead of fabricating it.
8. Add the author-bio/credential block for E-E-A-T, matching the voice identified in step 2.
9. Add the machine-readable layer if the piece warrants it. For a blog post or page that will be shared or indexed, hand over the slug + canonical URL, Open Graph / Twitter Card tags, and a JSON-LD Article block — see references/seo-packaging.md for the exact snippets and rules (real dates/author only; institutional pieces use an Organization author).
10. Run every new text you wrote (title, meta, headings, anchors, bio) through auditing-brand-voice's de-AI-flavor pass. Do not run the original draft body through it — this skill doesn't touch that.
11. Return the bundle as: title / meta description / section-title map / internal links / (press note if applicable) / author-bio block / (machine-readable layer if produced). Nothing else changes in the draft.

Common Mistakes

Related Skills

If this project doesn't yet have brand-kit reference material (fact sheet, voice-card, media-coverage inventory), run building-brand-kits first to produce it. For a voice/register/tone check on the original draft prose (out of scope for this skill), use auditing-brand-voice. For a full press-repurposing package beyond the short note in step 7, use repurposing-press-coverage. For making a page crawlable, renderable, and citable by AI answer engines (llms.txt, AI-crawler robots rules, citation-friendly structure), use web-quality:optimizing-for-ai-search; for auditing a live site's SEO/Core Web Vitals, accessibility, and crawlability end to end, use web-quality:auditing-website-quality.