sergekostenchuk

Use when admin/CMS/back-office screens need to be built or specified after admin-ui-orchestrator defines roles, permissions, workflows, and safety gates. Covers tables, forms, editors, moderation queues, dashboards, CRUD surfaces, and state coverage without production destructive actions.

Use when a project needs admin/CMS/back-office UI architecture: roles, permissions, content workflows, AI assistant boundaries, audit logs, destructive-action confirmations, and builder handoff. Planning only; do not perform production admin actions or expose secrets.

Use this skill when the user wants a human-readable project progress screen, agent work visualization, customer-facing status cockpit, immediate bootstrap progress before TASK-PLAN.md exists, task-plan-derived HTML progress page, user decision gates, or auto-updating UI/UX project status for senior-ui-ux-orchestrator workflows. It renders and validates progress views from project-request.json, workflow-log.jsonl, TASK-PLAN v2, progress-state.json, feedback artifacts, reports, and wiki decisions without becoming the source of truth.

Use when a website or web app needs privacy-aware analytics architecture: GA4, Yandex Metrica, self-hosted analytics, consent, UTM, event taxonomy, funnels, dashboards, and launch verification. Do not mutate real analytics accounts or store credentials without explicit approval.

Validate proposed or existing backlinks for relevance, editorial legitimacy, link attributes, spam/toxicity risk, platform compliance, anchor quality, target-page fit, and monitoring evidence. Use this skill when the user asks whether an external link opportunity or placed backlink is safe and useful, and reject toxic, irrelevant, paid-undisclosed, fake, or spam placements.

Turn semantic core, business constraints, and user feedback into a site concept and throwaway wireframe prototype before visual design. Use when a public site project needs a pre-design concept, content hierarchy, concept conflicts, or business-idea feedback before Stitch, Pencil, Figma, or code.

Use when planning deployments, build commands, environment checks, CI/CD, release steps, smoke tests, rollback, and monitoring. Dry-run first; do not deploy to production, change secrets, or mutate live services without explicit approval.

Collect, structure, deduplicate, and resolve user and agent feedback on UI/UX skills, subprocesses, design directions, progress-screen gates, Stitch/Pencil/Figma outputs, and implementation steps. Use when comments need statuses such as open, accepted, partially accepted, applied, rejected, or when review-board/progress-page fields must become actionable decisions.

Coordinate SEO, LLM-readable site architecture, UX, content, monitoring, security, and external authority work as a task-plan-driven sequence. Use this skill when the user asks to plan, execute, validate, or route multi-skill website growth work; asks what to do next across semantic core, URL structure, internal linking, schema, llms.txt, UX journeys, analytics, citation monitoring, or white-hat link placement; or wants a central architect that preserves evidence, alarms, handoffs, and tests.

Validate proposed or existing backlinks for relevance, editorial legitimacy, link attributes, spam/toxicity risk, platform compliance, anchor quality, target-page fit, and monitoring evidence. Use this skill when the user asks whether an external link opportunity or placed backlink is safe and useful, and reject toxic, irrelevant, paid-undisclosed, fake, or spam placements.

Review public site content for editorial quality, source support, prompt residue, duplicate blocks, translation drift, hidden filler, and human usefulness before SEO or LLM-friendly publication. Use this skill when the user asks to quality-gate news briefs, longform explainers, blog posts, topic pages, project pages, translated content, or generated article drafts without changing the live publishing pipeline.

Find and score legitimate external authority and backlink opportunities as a dry-run register, using relevance, platform rules, risk, target-page fit, and explicit user approval gates. Use this skill when the user asks where links to a target site could be placed effectively, but never post, submit, DM, email, open PRs, or edit external accounts without separate explicit approval.

Convert a semantic core into crawlable SEO and LLM-friendly information architecture. Use this skill when the user asks for URL structure, page taxonomy, canonical rules, hreflang groups, page roles, pillar/topic/news/article/blog/project separation, multilingual route strategy, or a durable content model such as one short brief plus one longform article. It prevents duplicate public content bodies and hands link-graph work to internal-link-graph-architect.

Design intentional internal link graphs for SEO, UX, and LLM-readable sites. Use this skill when the user asks for links between short briefs, longform articles, topics, projects, author pages, breadcrumbs, source trails, related stories, orphan-page checks, overlink checks, anchor intent, or validation that a link graph reinforces the canonical content model without creating duplicate story bodies or hidden links.

Plan and record evidence-backed LLM and AI-search citation checks across approved assistant/search surfaces with timestamped queries, locales, modes, cited URLs, snippets, screenshots or transcripts, and caveats. Use this skill when the user asks whether ChatGPT, Perplexity, Claude, Gemini, Copilot, or another assistant cites a target site, but do not claim citation without direct evidence.

Design human-visible, machine-readable website layers for AI assistants, RAG, and agentic search. Use this skill when the user asks for llms.txt, LLM discovery plans, visible TLDR or direct-answer blocks, source trails, entity pages, markdown alternates, AI crawler policy, topic-to-URL maps, assistant-readable public content, or LLM-friendly architecture that must preserve canonical SEO strategy without hidden bot-only facts or duplicate story bodies.

Use this senior orchestrator whenever an agent must analyze, route, review, or draft strategy for Russian traffic-fine appeals in ANTIPENALTY. This skill coordinates traffic-case-router, traffic-evidence-auditor, traffic-court-router, traffic-admin-defense-lawyer, traffic-complaint-strategist, traffic-complaint-drafter, and legal-fail-closed-reviewer. Use it for multi-step legal defense workflows, case analysis, complaint generation planning, role-confusion risks, missing evidence, or any request where a traffic penalty must not become a blind auto-complaint.

Use this skill whenever ANTIPENALTY must review a traffic-fine strategy, complaint draft, court route, evidence audit, or legal claim before generation or user delivery. This reviewer blocks hallucinated courts, unsupported facts, role confusion, weak arguments promoted as strong, missing evidence, and legal claims without source support. Use it as the final gate after strategist or drafter output and whenever the user asks whether a legal output is safe.

Use this skill whenever ANTIPENALTY must analyze legal defenses for Russian traffic administrative penalties, including GIBDD/CAPAF auto-camera cases, MADI parking/standing cases, AMPP paid parking cases, camera verification, owner-not-driver arguments, photo quality, address errors, and evidence requests. This skill ranks defense lines by strength, requires evidence, marks weak or dangerous arguments as do_not_use, and must run after traffic-case-router and traffic-evidence-auditor.

Use this skill whenever ANTIPENALTY must classify a Russian traffic fine, постановление, protocol, parking fine, evacuation case, accident case, or authority document before legal analysis. This skill determines case_source and issuing_authority, flags missing facts, and blocks wrong playbooks such as applying GIBDD camera logic to MADI or AMPP cases. Use it before any evidence audit, defense strategy, court routing, or complaint drafting.

Use this skill whenever ANTIPENALTY must draft traffic-fine complaint text from an already confirmed strategy, confirmed facts, court/addressee route, and evidence list. This skill must not invent facts, legal grounds, courts, dates, drivers, camera defects, or evidence. Use it only after traffic-complaint-strategist and before legal-fail-closed-reviewer.

Use this skill whenever ANTIPENALTY must turn a traffic fine evidence audit and legal defense analysis into a focused complaint strategy. This skill selects 1-3 supported defense lines, keeps weak theories out, creates questions and motions, and decides whether generation is blocked. Use it after traffic-admin-defense-lawyer and traffic-court-router, before any complaint drafting.

Use this skill whenever ANTIPENALTY must decide where a Russian traffic-fine complaint should be filed or whether court/addressee routing must remain unresolved. This skill is mandatory before complaint drafting when a court, authority, address, jurisdiction, MADI/AMPP route, GIBDD route, or place-of-violation issue appears. It prevents invented courts and returns fail-closed unresolved output when location or routing source is insufficient.

Use this skill whenever ANTIPENALTY must audit evidence, extracted OCR/PDF facts, uploaded documents, user profile data, user statements, or missing proof before choosing a traffic-fine defense strategy. This skill separates confirmed facts from assumptions, detects owner/applicant/driver/postal-recipient/account-holder confusion, identifies evidence gaps, and blocks unsupported claims about cameras, verification, another driver, court routing, or procedural defects.

Create, improve, evaluate, harden, and package production-grade Codex/Claude skills and skill groups. Use when the user wants to turn a workflow into a reusable skill, upgrade an existing skill, create multiple coordinated skills, design orchestrator-worker systems, run skill evals or benchmarks, verify tool/MCP dependencies, harden dependency install policy, optimize a skill description for triggering, or produce an evidence-backed delivery report. Also use for lightweight "vibe" skill drafting when the user explicitly wants less process.

Demonstrate false positive lint behavior for failure retry evidence terms without linked production references.

Create a small verified artifact from a user workflow. Use when the user asks to capture a workflow as a reusable skill with evidence, tool checks, failure modes, retry limits, and final review.

Demonstrate missing reference validation for a skill with production gates and evidence requirements.

Help write better changelogs.

Experimental NotebookLM control path for Firefox using WebDriver BiDi or geckodriver, with verification-first gates and no CDP assumptions.

Operate the logged-in NotebookLM UI through a visible Chromium/CDP browser session, preferring Comet and falling back to Chrome, Brave, Edge, Vivaldi, Opera, or Arc for source collection, Fast/Deep Research, YouTube/web URL import, chat synthesis, and Studio artifacts.

Experimental NotebookLM control path for Safari using safaridriver/WebDriver, with verification-first gates and no CDP assumptions.

Run a mandatory cleanup pass on files touched by the current implementation task before it can close. Use this skill whenever a task is moving from implementation/review/testing toward final done and there is any chance of dead helpers, unused imports, commented-out code, stale flags, obsolete wrappers, or duplicate transitional logic remaining in the touched files.

Run a deep, manual, project-wide audit for dead code, duplicated logic, stale wrappers, and legacy residue before major refactors or during periodic maintenance. Use this skill whenever a repository needs a human-reviewable cleanup plan across the whole codebase, especially when old architectural layers, abandoned adapters, or oversized legacy files may still be present.

Build, repair, tune, and validate interactive cursor-trail mask reveal hero effects in frontend projects. Use when implementing canvas-based base/reveal compositing, soft radial cursor masks, fading cursor trails, reveal debug controls, full-screen hero scenes without site text, or browser smoke checks for cursor reveal, hover reveal, before/after reveal, and mask compositing effects.

Validate whether two raster image layers are suitable for reveal, before/after, morph, mask-compositing, or interactive cursor reveal effects. Use when comparing base/reveal images, checking whether the main subject stayed in the same position, separating primary subject drift from background or secondary-object differences, producing annotated overlays/difference maps, or deciding whether an image layer must be shifted, scaled, cropped, or regenerated before frontend compositing.

Coordinate the local Xray/VLESS/XHTTP VPN workflow, route tasks to internal worker modules, enforce scoped alarms, and accept outputs only after evidence and review.

Recommend UI pattern sources, component registries, visual references, and motion/runtime options for UI/UX projects with stack compatibility, freshness labels, no-copy boundaries, evidence manifests, and handoffs to design, build, critic, and progress-dashboard skills. Use when a user asks what UI libraries, references, patterns, effects, dashboards, admin components, or inspiration sources should inform a site, app, admin panel, or agent interface.

Audit Codex and compatible filesystem skill portfolios for exact or partial duplication, unclear boundaries, stale variants, and orchestrator-worker opportunities. Use when a user wants to inventory one or more provider skill directories, compare changes, review overlapping skills, maintain an evidence-backed skills wiki, or prepare safe consolidation proposals without automatically modifying runtime skills.