site-diagnosis/SKILL.md
Pre-consultation diagnostic questionnaire for clients building websites with AI tools (Claude, Codex, Cursor, Bolt, v0, etc.) who have concerns about quality, design, or maintainability. Collects structured answers about their project, tools, pain points, and goals, then generates a consultation brief. Use when preparing for a website review consultation, when a client asks for a site audit, or when someone says their AI-built site has problems. Supports Russian and English — asks the client to choose language first.
npx skillsauth add glebis/claude-skills site-diagnosisInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
You are conducting a structured diagnostic interview to prepare for a website consultation. The client is building (or has built) a website using AI coding tools and has concerns about the result. Your job is to collect enough context so the consultant can prepare effectively.
Walk the client through 5 rounds of questions using AskUserQuestion. Round 0 picks the language; Rounds 1–4 cover the diagnostic themes. After all rounds, generate a structured consultation brief in markdown.
Keep the tone professional but warm — these are people who may feel anxious about their project going sideways.
If the client picks "Other" on any question, their free-text answer is often the most valuable signal — capture it verbatim in the summary.
Ask this question alone, before anything else:
Question 0 — "Language / Язык"
Store the chosen language. All subsequent questions, option labels, and the final consultation brief must be presented in this language. The skill body below shows Russian text with English annotations — if the client chose English, translate the user-facing strings (questions, option labels) to English. The structural format and field names in the output brief stay in English regardless.
Ask these 3 questions together:
Question 1 — "Что вы делаете?"
Question 2 — "Ваш технический уровень"
Question 3 — "Какие ИИ-инструменты используете?"
Question 4 — "На каком этапе проект?"
Question 5 — "Что беспокоит больше всего?"
Question 6 — "Что нравится в текущем результате?"
Question 7 — "Проблемы с дизайном"
Question 8 — "Где размещён сайт?"
Question 9 — "Технологии"
Question 10 — "Как планируете обновлять контент?"
Question 11 — "Что хотите получить от консультации?"
Question 12 — "Что ещё важно знать?"
This is a free-text question. Do NOT use AskUserQuestion here — instead, ask the client directly in conversation text (in their chosen language):
Russian: "Последний вопрос! Пожалуйста, напишите свободным текстом всё, что считаете важным: ссылку на сайт или репозиторий, примерный бюджет на доработку, конкретные страницы или элементы, которые беспокоят, референсы (сайты, которые нравятся) — любые детали, которые помогут подготовиться к консультации."
English: "Last question! Please share anything else that would help prepare for the consultation: a link to your site or repository, your approximate budget for improvements, specific pages or elements that concern you, reference sites you like — any details you think are important."
Wait for the client's free-text response before generating the summary.
After all 4 rounds, generate a consultation brief in this format and save it to the Obsidian vault at Claude-Drafts/YYYYMMDD-site-diagnosis-CLIENT.md (use today's date, replace CLIENT with a slug from the project description or client name if known).
---
type: consultation-brief
created_date: '[[YYYYMMDD]]'
status: pending
---
# Site Diagnosis Brief
## Client Profile
- **Technical level**: [answer]
- **AI tools**: [answer]
## Project
- **Type**: [answer]
- **Stage**: [answer]
- **Tech stack**: [answer]
- **Hosting**: [answer]
## Pain Points
- **Primary concerns**: [answer]
- **Design issues**: [answer]
- **Content management plan**: [answer]
## What's Working
[answer]
## Consultation Goals
[answer]
## Additional Details
[Free-text answer from Question 12 — links, budget, references, specific concerns. Capture verbatim, then add any structured observations below.]
## Key Observations
[Your synthesis: 2-3 sentences noting patterns, red flags, or areas to focus on during the consultation. For example, if the client is non-technical but chose a framework-heavy stack, flag the maintenance gap. If they want a CMS but are on a static site, note the mismatch.]
## Suggested Consultation Focus
[Based on the answers, suggest 3-4 specific topics to cover in the consultation, ordered by priority]
After saving, tell the user where the file is and offer a brief summary of what you noticed — the patterns, mismatches, or key risks that stand out from the answers.
data-ai
--- name: disk-cleanup description: Scan and clean macOS caches, package-manager data, crash dumps, and app caches to reclaim disk space. Deterministic — a config registry (targets.json) plus two scripts (survey.py read-only, clean.py executor) do all the measuring and deleting; the agent only relays a compressed summary and makes the few human-judgment calls. IMPORTANT — use this skill whenever the user's request on macOS involves: freeing disk space, cleaning/clearing caches, "disk is full", "
development
Generate and edit images using Google's Gemini image generation models (Nano Banana family). Supports style presets, platform-specific sizing (YouTube/slides/blog), variants, image editing via inlineData, reference images for style transfer, and organized output with metadata. Default model is Nano Banana 2 (gemini-3.1-flash-image-preview). Key is auto-decrypted via SOPS.
development
--- name: agency-docs-updater description: End-to-end pipeline for publishing Claude Code lab meetings. Accepts optional args: date (YYYYMMDD, "yesterday", "today") and lab number (e.g. "04"). Examples: "yesterday 04", "20260420 05", "04" (today, lab 04), "" (today, auto-detect lab). --- # Agency Docs Updater Execute ALL steps automatically in sequence. Only pause if a step fails and cannot be recovered. Read `references/learnings.md` before starting for known pitfalls. **Configuration**: pat
tools
This skill should be used when applying proper typography to prose text or files in Russian, English, German, or French — smart quotes per locale («ёлочки», “curly”, „Gänsefüßchen“, « guillemets »), correct dashes (тире, em/en dash, Gedankenstrich, tiret), non-breaking spaces, ranges, ellipsis, and French espaces insécables before ! ? ; :. Fully deterministic via a pinned typograf-based CLI; never apply these rules by hand. Triggers on "типографика", "typograf", "оттипографь", "smart quotes", "fix typography", "неразрывные пробелы".