Fully agentic portfolio creation for your AI Code: a state-aware router, eight full skill packs, graph-aligned subagents, deterministic build/QA tooling, hooks, bilingual RTL support, and approval-gated publishing.
Skills 8 in the Claude listing
Portfolio audit Use to review an existing generated portfolio for factual provenance, privacy, accessibility, responsive behavior, Arabic RTL, broken links, rendering, SEO/GEO metadata and release regressions. Portfolio build Use when approved portfolio facts, copy and layout must be generated or regenerated as the repository's static website. Portfolio design Use for portfolio information architecture, section hierarchy, responsive visual direction, design tokens, bilingual/RTL behavior, accessibility and UX repair once messaging is approved. Portfolio interview Use when the portfolio workflow needs direct user input that cannot be recovered safely from approved sources: missing facts, target reader, positioning, language/register, visual preferences, confide Portfolio publish Use only when the user explicitly requests publishing or deploying a reviewed portfolio to an approved destination. Portfolio router Use as the entry skill for ambiguous, multi-step, repair-heavy or approval-gated portfolio requests. Portfolio source Use when portfolio facts must be extracted or verified from a CV, resume, LinkedIn export, GitHub profile, old website, case-study files, testimonials, analytics or other supplied sources. Portfolio story Use to write, rewrite or localize approved portfolio headlines, summaries, About sections, project case studies, services and SEO metadata in English or Arabic. 