Alena Zakharova's skills
Share
Claude Code skills for product and UX work: website UX audit, requirement docs, a PMF cycle, text clustering and ADRs
Install
Download https://github.com/alenazaharovaux/share and copy the needed skill folders from skills/ into ~/.claude/skills/, for example ux-audit, req-docs, pmf and topic-clustering.This is third-party code. Review the repository files before installing.
What it does
The repository collects Claude Code skills; some are written by the author and some are adapted from open projects with licenses credited. The author's skills run a 50-item prioritized website UX audit, choose and write requirement documents from 13 templates, and guide a product through ten stages of finding product-market fit. The topic-clustering skill sorts large text sets, such as a Telegram export, into topics with BERTopic. There are also skills for an architecture decision log, HTML slides and prompts for image generators.
Who it is for. For product managers, UX researchers, founders and writers using Claude Code in Russian or English.
Good fit when
- You need a structured, prioritized UX audit of a landing page
- You need to decide which requirement documents a project needs and write them
- You want to validate a product idea step by step and resume weeks later
- You need to sort a large chat export or transcripts into topics
Not a fit when
- You want one-command or marketplace install; skills are copied by hand
- You want only original work: the journalism skills and some HTML skills come from other repositories
Example request
Run a UX audit of example.com and give me a prioritized report on what hurts conversionLimitations
Skills install by copying folders; there is no package or marketplace for the set. Some skills are adapted from superpowers, jamditis/claude-skills-journalism, nicobailon/visual-explainer and official Anthropic plugins. The UX audit needs a page fetching tool: built-in WebFetch or a search MCP. topic-clustering installs Python packages and runs a script. Some instructions, for example in adr, assume the author's Obsidian workflow.
How to disable. Delete the skill's folder from ~/.claude/skills.
Security check
- topic-clustering installs Python packages and runs a script
- The UX audit fetches pages from the web
README in short
The README, in English and Russian, groups the skills into writing, HTML pages and presentations, and development workflow. Each entry names its author or source and license. Installation is copying a skill folder into ~/.claude/skills. It also mentions the claude-doctor plugin, installed via marketplace from a separate repository, and a prompt set for analyzing diary studies.
SKILL.md
--- name: ux-audit description: > Run a full UX audit on any website: Nielsen heuristics, conversion, content, technical quality, information architecture. Produces a prioritized report with evidence-based findings and actionable recommendations. Use when asked to review a site, check a landing page, find UX problems, evaluate usability, assess conversion, or anything like "what's wrong with this site", "review the website", "audit UX", "check the forms", "why isn't the site converting". --- # UX Audit — full website audit You run an expert UX audit. Your job is not just to list problems — it's to explain **why** each one hurts the business and **what exactly** to do about it. The audit client is usually a business owner or marketer, not a UX specialist. Argue in business terms (conversion, trust, lost customers), not design patterns. **Language:** Communicate in the same language the user speaks. Write the report in that language too. ## Prerequisites — web crawling tool This skill needs a way to fetch website content. The author tested it with [Exa MCP server](https://docs.exa.ai/reference/mcp) (`web_search_exa` with `livecrawl: "preferred"`), which gives the best results for page content extraction. If you don't have Exa, you can use: - **WebFetch** — built into Claude Code, no setup needed - **WebSearch** — built into Claude Code, works for basic page analysis - **Other MCP search servers** — any tool that can fetch page content
FAQ
Which skills are the author's own?
ux-audit, req-docs, pmf, topic-clustering, adr, frontend-slides and illustration-prompt. The rest are adapted from other open repositories, with sources credited in the README.
Do they work in Russian?
pmf and topic-clustering are bilingual, ux-audit answers in the user's language, and adr recognizes Russian triggers.
Related
A CLI for every Google Workspace API with JSON output and agent skills: Drive, Gmail, Calendar, Sheets and more
Local search over Markdown notes, docs and meeting transcripts: keywords, semantic search and reranking, with an MCP server
A task manager for AI-driven development: breaks a PRD into dependent tasks and guides the agent through them via MCP or CLI
Anthropic's official plugins for knowledge work roles: sales, support, product, marketing, legal, finance, data and enterprise search