ng-galien

Orientation skill for Maket visual document work. Use when the user asks Claude to design, edit, review, or learn Maket; call maket_learn first for live product guidance, then operate through MCP tools.

Orientation skill for Maket visual document work. Use when Codex is asked to design, edit, review, or learn Maket; call maket_learn first for live product guidance, then operate through MCP tools.

Review and fix Maket documents — audit HTML quality, charte compliance, image paths, layout issues, and automatically apply corrections with maket_html patch. Use this skill when the user says "review", "check", "fix", "audit", "correct", "improve" a document, or when they report visual problems like broken images, wrong fonts, bad spacing, or charte violations. Also triggers when the user opens an existing document and wants it cleaned up or brought up to standard.

Brand-identity expert for Maket. Builds professional design-token systems (colors, type, spacing) from a brief, an industry, a website URL, or existing imagery. Researches current references on the web. Triggers when the user wants to create a charte, define a visual identity, pick colors/fonts for a project, or says "make me a brand guide", "choose the colors", "what typography for...", "create the visual identity". Also when the user mentions a sector (restaurant, real estate, yoga, tech) and wants a matching style.

Explore and diagnose any codebase through code-moniker's symbolic index, using the best installed surface (MCP tools when configured, otherwise the local binary). ALWAYS use this before grep/Glob/Read exploration when the question involves architecture, module structure, coupling or dependencies between parts, call graphs, callers/callees, impact of a change, code smells, refactor targets, or codebase health — manual exploration produces impressions, code-moniker produces counted facts (coupling edges with counts, ranked violations) in fewer calls. Typical requests, in any language: map the architecture, architecture or module overview, dependency graph, strongest or heaviest couplings, who calls or uses X, impact of changing X, where to refactor, is this codebase healthy, review this diff structurally. Zero project configuration on any ts/rs/java/python/go/cs/sql project.

Review a repository with code-moniker's local check DSL for Fowler and Lanza-Marinescu code smells. Use when Codex needs to write, validate, or run project-specific warning-severity smell rules, distinguish executable local DSL checks from AST/corpus/history checks, triage check output, or plan follow-up evolutions for missing smell-detection operators.