Help us improve
Share bugs, ideas, or general feedback.
From agentkit-seo
Optimizes personal websites and portfolios for search discoverability, crawlability, structured data, metadata, and AI-readable signals. Guides audits and improvements for SEO and web-based personal branding.
npx claudepluginhub agentkit-seo/agentkit-seo --plugin agentkit-seoHow this skill is triggered — by the user, by Claude, or both
Slash command
/agentkit-seo:agentkit-seo-web-portfolioThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Work through the lens of a technical SEO specialist and a hiring manager skimming the site. Use this skill to improve how a personal site is crawled, rendered, summarized, and trusted by search engines and AI systems.
Audits web codebases for SEO and AI-answer visibility, checking page speed, crawlability, semantics, headings, then generates prioritized fix-it plans.
Optimizes websites for search engines via technical SEO: robots.txt, meta robots, canonical URLs, XML sitemaps, URL structures, per Lighthouse audits and Google guidelines.
Scans websites for SEO issues including meta tags, headings, broken links, images, Core Web Vitals, and structured data, producing a prioritized fix list.
Share bugs, ideas, or general feedback.
Work through the lens of a technical SEO specialist and a hiring manager skimming the site. Use this skill to improve how a personal site is crawled, rendered, summarized, and trusted by search engines and AI systems.
llms.txt copy: references/section-recipes.mdllms.txt, AI retrieval, evidence labels, source confidence, platform constraints, or known agent failure modes.references/ file and mark wiki-specific guidance as unavailable when it affects confidence.Use the smallest honest audit depth:
Quick scan: homepage, robots, sitemap, title/meta/canonical basics, main navigation, and visible positioning.Default audit: quick scan plus up to 2 user-specified or visibly priority pages, structured data, Open Graph, internal links, and top project pages when available.Deep audit: full route inventory, built HTML/source templates, performance/mobile checks, redirects/status codes, schema validation, broken links, and code edits.Default to Default audit for broad portfolio audits. Offer Deep audit as an optional next step when the current answer would benefit from more evidence. Do not choose Deep audit silently unless the user asks for a full site audit, exact code changes, launch validation, or every important route checked.
llms.txt.Before returning, check the draft and fix or flag any failure:
If a check fails and cannot be fixed from available inputs, say so rather than papering over it.
Return:
For audits, use concise labels such as Verified, From source, From context, Inference, and Inaccessible when a claim could otherwise be ambiguous. Mark unsupported responsibilities, metrics, seniority, clients, testimonials, or outcomes as gaps rather than turning them into metadata, schema, or copy. When the audit is intentionally bounded, include a one-line Depth note that says what was not inspected and what deeper inspection would add.
When the user asks for a score, scorecard, or before/after comparison, also apply references/audit-scoring.md: report the overall score, band, per-category breakdown, and a fix-first ranking, labeled as an internal prioritization heuristic rather than a ranking or indexing guarantee.
Human playbook: hub/web-portfolio/README.md.