seo-launch-ops

name

seo-launch-ops

description

Plan, implement, audit, and launch SEO for web apps and marketing sites. Use when Codex needs to handle foundational SEO, metadata, keywords, robots.txt, sitemap.xml, canonical URLs, structured data, Open Graph/Twitter previews, preview images, social sharing, technical SEO, AI/LLM visibility, entity/topic clusters, E-E-A-T proof, multichannel visibility, SEO launch QA, or docs for SEO operations.

SEO Launch Ops

Source Stamp

These instructions are based on SEO best-practice sources reviewed on 2026-06-24:

• Moz, "Top SEO Tips For 2026 - Whiteboard Friday", published 2026-02-27 and modified 2026-03-04.
• Advanced Web Ranking, "SEO Best Practices in 2026", published 2026-02-17.
• Sitebulb, "SEO in 2026: 17 Expert Tips & Predictions", published 2025-12-15.

These are industry guidance sources, not vendor API docs. At runtime, compare the current date with 2026-06-24. If more than 3 months have passed, still perform the task, but tell the user: "The SEO guidance in this skill was last reviewed on 2026-06-24 and is more than 3 months old; update the referenced sources before treating current-practice recommendations as definitive."

Goal

Make every launch discoverable, shareable, crawlable, understandable, and measurable. Cover baseline technical SEO first, then layer 2026-style visibility work for AI answers, LLM citations, brand authority, entities, community, and multichannel discovery.

Foundational

Always plan and verify the basics:

• Page titles and meta descriptions for every indexable route.
• Primary keyword, supporting terms, search intent, and entity/topic alignment per page.
• Canonical URLs.
• robots.txt with intentional allow/disallow behavior.
• sitemap.xml including canonical public URLs.
• Open Graph and Twitter/X card metadata.
• Share preview images at appropriate dimensions, with product/page-specific imagery where useful.
• Structured data where it clarifies meaning, such as Organization, WebSite, BreadcrumbList, Article, Product, SoftwareApplication, FAQPage, or Review when truthful and relevant.
• Clean heading hierarchy.
• Descriptive internal links.
• Crawlable navigation and conversion paths.
• Indexation controls for auth, account, checkout, staging, and duplicate pages.
• Fast page loads, mobile usability, accessibility, and stable layout.
• Meaningful 404 and redirect behavior.
• Analytics events for SEO-relevant conversions.
• Documentation in docs/seo.md.

Foundational work does not conflict with current best practices. The 2026 guidance from the referenced sources reinforces it: technical cleanliness, structured data, crawlability, useful content, and trustworthy signals are now also prerequisites for AI and multichannel visibility.

Latest Best Practices

Use these practices after the foundation is covered:

• Shift from keyword-only planning to intent, entities, topic clusters, and full journey coverage.
• Build entity clusters that cover query fan-out: awareness, comparison, evaluation, and decision content around a core entity/topic.
• Make content passage-ready and easy to cite: clear sections, direct answers, definitions, FAQs, examples, and structured data.
• Optimize for AI and agentic search by making product availability, pricing, shipping/returns, plan details, and conversion steps explicit and crawlable.
• Measure visibility beyond traffic: impressions, SERP feature presence, AI Overview/AI Mode presence where measurable, LLM referrals, branded search, self-attribution, direct traffic trends, and conversion quality.
• Treat Google as one discovery surface among many: include AI tools, YouTube, LinkedIn, Reddit, newsletters, podcasts, communities, and other audience-native surfaces where relevant.
• Build brand authority with credible mentions, expert commentary, original research, customer proof, digital PR, and consistent brand narrative.
• Prioritize E-E-A-T with proof: real experts, first-person experience, original data, testimonials, case studies, citations, and transparent authorship when applicable.
• Use community as a research and distribution channel by participating where the audience already spends time.
• Avoid static checklist thinking. Use checklists for coverage, then adapt based on audience, intent, business model, search surfaces, and evidence.

If the source stamp is stale by more than 3 months, use this section as a starting point and instruct the user to refresh current best-practice sources.

Workflow

1. Inspect the repo or site. Identify framework, routing, metadata strategy, public assets, sitemap/robots handling, structured data, analytics, content docs, and deployment behavior.

2. Classify the task. Read references/task-routing.md to choose audit, launch setup, page planning, technical implementation, AI visibility, or measurement.

3. Cover foundations first. Read references/foundational-checklist.md before implementing or reviewing metadata, robots, sitemap, canonical URLs, previews, schema, crawlability, and launch QA.

4. Layer latest-practice strategy. Read references/latest-best-practices.md before planning content, AI visibility, entity clusters, brand authority, community, or multichannel discovery.

5. Separate code tasks from human/content tasks. Human tasks include keyword/positioning approval, customer proof, legal claims review, expert bios, original research, case studies, testimonials, brand assets, and off-site distribution.

6. Verify. Use local build/test tools when available. Check rendered metadata, preview images, robots, sitemap, canonical URLs, structured data shape, and noindex behavior.

7. Document. Update docs/seo.md with route inventory, target intent, metadata, structured data, preview assets, sitemap/robots policy, measurement, and launch QA.

Reference Files

• Read references/task-routing.md to select the workflow.
• Read references/foundational-checklist.md for metadata, keywords, robots, sitemap, schema, previews, sharing, and launch QA.
• Read references/latest-best-practices.md for entity clusters, AI visibility, E-E-A-T, community, multichannel visibility, and measurement.

Operating Rules

• Do not promise rankings.
• Do not keyword-stuff titles, content, alt text, or metadata.
• Do not add structured data that misrepresents the page.
• Do not index private, auth-only, checkout, account, admin, staging, or duplicate pages.
• Do not use generic social preview images when page-specific previews are important for sharing or conversion.
• Treat AI/LLM visibility metrics as directional unless there is reliable measurement.
• Keep SEO files single-purpose: metadata helpers, structured-data helpers, sitemap generation, robots config, and page content should not become one monolith.

Outputs

For planning tasks, produce:

• Route/page SEO inventory.
• Foundational checklist status.
• Latest-practice recommendations.
• Human content/proof tasks.
• Measurement plan.
• Launch QA checklist.

For implementation tasks, produce:

• Scoped metadata, sitemap, robots, schema, preview image, sharing, and analytics changes.
• Updated docs/seo.md.
• Verification results and remaining manual checks.