kyzdes

Generate UI designs with Google Stitch AI. Use when user asks to create UI, design a page, prototype an interface, generate a screen, make a mockup, or mentions stitch or design prototyping. Conversational brainstorm-first workflow.

Edit an existing Stitch UI design. Use when user asks to modify, change, update, or refine an existing screen design such as changing headers, adding sidebars, or adjusting colors.

Upload a wireframe, screenshot, or sketch to Google Stitch for digitization into a clean UI. Use when user provides an image file and asks to convert it to UI or digitize a sketch.

Extract design theme from a Stitch-generated HTML file. Use when user asks to extract Tailwind config, fonts, icons, or design tokens from a generated design.

Deploy and manage applications on VPS servers with Dokploy. Use when the user wants to: set up a new VPS server, deploy a project from GitHub, manage domains/DNS, create databases, check server status, view logs, or remove deployed projects. Also use when the user mentions re-deploying, checking deploy status, adding environment variables, or troubleshooting a deployed app. Also triggers when users say things like "put this on my server", "I need hosting", "make this accessible online", "my site is down", "set up CI/CD for deployment", or anything related to getting code running on a remote server. Triggers on: VPS, deploy, server setup, Dokploy, hosting, domain, DNS, redeploy, server status, deploy logs, "put online", "host this", "site down", CI/CD.

Securely save, retrieve, or reference API keys, SSH keys, server credentials, and domain info via the `keys` CLI, without ever exposing the plaintext value. Use when the user mentions saving, getting, injecting, or referencing secrets, API keys, tokens, SSH keys, server addresses, or domain configs.

Turn loose product/feature descriptions in any language into a structured English UX Spec ready to hand off to a visual-design agent (huashu-design). Triggers on Russian and English product ideas — "хочу приложение для X", "сделай продукт", "нужен dashboard", "design a SaaS", "make me a dashboard", "I want an app for", "feature spec for", "UX spec" — especially when the description is vague, incomplete, or product-shape is unclear. Switches to advisor mode (proposes 2–3 archetypes with tradeoffs) when the user is stuck. Output spec is always English regardless of input. Make sure to use this skill whenever the user describes WHAT they want to build before going to design — the spec pre-answers huashu's 10 standard questions and prevents generic output. NOT for visual design execution, decks, animations, infographics, PRDs, or market research — those go elsewhere.

Manage HostBRR VPS via the VirtFusion REST API — list servers, read specs/IPs, rebuild/reinstall the OS, power actions (restart/boot/shutdown/powerOff), SSH keys, ISO, VNC, rescue mode, resource packs, and task polling. Use whenever a task involves HostBRR servers, vps.hostbrr.com, "rebuild/reinstall a VPS", or a HostBRR API token. Teaches the correct client API base path (/api, NOT /api/v1), Bearer auth kept out of the transcript, the FQDN-hostname rebuild rule, and async task polling.

Work the Agentix issue tracker — an agent-native, Linear-style tracker (projects, epics, issues, milestones, checklists, wiki) reached over a REMOTE, hosted MCP server named `agentix` (REST mirror at `/api/*`). Use whenever the agentix MCP tools are present (get_started, get_context, create_issue, search, create_document, set_task_spec…), OR the user mentions Agentix / "the tracker" / "заведи/создай задачу" / "create an issue/task" / "log work" / planning epics/milestones / the wiki/index/"оглавление", OR asks to write, file, sync, or push tasks, specs, backlog, epics, context, or docs INTO Agentix — even from another project or repo. Agentix is a running service you connect to over MCP; a local agentix repo is the source code, NOT how you use it (never run it locally). Teaches finding the URL+token (incl. keys-keeper) and connecting, the orient workflow (get_started → index → get_context), the task template, and the index convention. NOT for GitHub Issues, Jira, or the hosted Linear MCP.

Build travel itineraries from Aviasales and Ostrovok — extract flight/hotel data and compile a self-contained HTML with XLSX/PDF export and optional Vercel deploy. Triggers on avs.io / aviasales.ru / corp.ostrovok.ru links (even pasted without explanation), on Russian travel terms (отпуск, поездка, перелёт, отель, бронирование, маршрут), and on planning a trip with dates/destinations. Also remembers previously planned trips across sessions in ~/.trip-planner/ (recall past trips first, record each when done), so it triggers on 'какие поездки я уже планировал', 'покажи прошлые маршруты', 'what trips have I planned'.

Instrument an app with SentryX — a self-hosted Sentry-analog that unifies error tracking, distributed tracing, and product analytics/funnels, correlated by trace_id — exposed over a remote MCP server named `sentryx`. Use whenever the sentryx MCP tools are present (get_started, whoami, create_project, instrument_hint, define_event/funnel/feature, get_funnel, search_issues, get_issue, get_trace, prepare_fix_bundle) or the user mentions SentryX, "instrument my app", "error tracking", "distributed tracing", "define a funnel", "product analytics", "trace debugging", "set up Sentry / SentryX", "OTLP", or "fix this bug with SentryX". Teaches the connect-once flow, the instrument flow (detect stack → 5 product questions → create_project → snippets → define_* → verify), and the fix-bugs flow (search_issues → prepare_fix_bundle). NOT for hosted Sentry.io, Datadog, or PostHog — those are different tools.

Build and maintain agent-facing project documentation in two linked layers: a COMMITTED `agent-docs/` navigation tree (a MAP router + per-domain deep docs + optional per-folder CLAUDE.md, so an agent reads only the slice it needs) and a GITIGNORED `context-map-<slug>/` memory tree (decisions, known issues, gotchas, tasks, Agent Conflict Protocol). Use when the user asks for a project context map, agent memory document, onboarding map, known-issues/decisions log; OR when a project is too big for an agent's context, agents get lost in the codebase, docs don't fit / waste tokens loading the wrong files, or the user wants to decompose / map / structure the project for AI, split docs by domain, set up per-area CLAUDE.md, or add a CI gate that keeps the docs fresh. Triggers in Russian too — "карта проекта", "память проекта для агента", "разложить проект", "агенты теряются в коде", "слишком большой проект", "документация для агентов", "decisions / known issues". Replaces the retired `agent-docs-architect` skill.

Decompose a large or complex codebase into agent-readable layered documentation — a top-level architecture map, per-domain deep docs, and distributed CLAUDE.md files in subfolders so agents read only the slice relevant to their current task. Use this whenever the user complains that "the project is too big for the agent's context", "agents get lost in the codebase", "docs don't fit", "agents waste tokens loading the wrong files", or asks to "decompose / map / structure / document the project for AI", "split docs by layer / domain", "set up per-area CLAUDE.md", "build a doc index for agents", "make the codebase agent-friendly", or considers RAG/LightRAG-style solutions for source-code documentation. Triggers on Russian phrasings too — "разложить проект", "документация для агентов", "агенты теряются в коде", "сделать карту проекта", "слишком большой проект". NOT a replacement for the `context-map` skill (which captures project memory — decisions/issues/gotchas); this is the layer above, focused on architectural