mcpField
E
existing-site-analyzer

existing-site-analyzer

github.com/sergekostenchukDesign & UI3251 installsUpdated 18w ago
Repository

About existing-site-analyzer

A skill that builds an evidence base for work on existing websites, intended for redesign and migration projects. It reviews local files, approved public pages, or screenshots and records content, constraints, assets, and risks without changing the live site.

Key features

  1. 01Inspect local pages and files
  2. 02Review public URLs with approval
  3. 03Analyze page screenshots
  4. 04Inventory media and brand assets
  5. 05Record evidence status and migration risks

Use cases

  1. 01Gather evidence for a website redesign
  2. 02Document page structure and SEO issues
  3. 03Inventory assets needed for migration
  4. 04Hand off findings to rebuild or migration planning

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/sergekostenchuk/ui-ux-agent-skill-system --skill existing-site-analyzer

After installing, type/existing-site-analyzerin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

agents/
openai.yaml
assets/
existing-site-audit.template.md
evals/
evals.json
references/
existing-site-analysis-contract.md

Details

From the project README

Existing Site Analyzer

This skill creates the evidence base for existing-site work. It does not decide rebuild strategy by itself and does not mutate the live site.

Operating Modes

  • local-inventory: inspect local HTML, source files, assets, sitemap, robots, and reports.
  • public-url-review: review public URLs or fetched HTML only after network/crawling approval.
  • screenshot-review: analyze screenshots when URL or source is unavailable.
  • asset-inventory: inventory images, videos, docs, icons, maps, and brand assets.
  • handoff: produce evidence for rebuild-or-improve-advisor and migration-planner.

Required Workflow

  1. Read references/existing-site-analysis-contract.md.
  2. Confirm source type and privacy:
    • local files and user-provided public screenshots: allowed;
    • public URL crawling/fetching: approval required if not already requested;
    • authenticated/private pages, analytics, CRM, or CMS: explicit approval and data-handling note required.
  3. Use assets/existing-site-audit.template.md for the report.
  4. Record evidence status for every source: Ran, Skipped, Planned, or Manual.
  5. Capture:
    • page inventory;
    • content and IA;
    • SEO/LLM surface;
    • UX and conversion issues;
    • technical constraints;
    • asset inventory;
    • migration risks.
  6. Hand off to rebuild-or-improve-advisor; involve migration-planner if URL structure, redirects, sitemap, robots, schema, or launch risk may change.

Safety Rules

  • Do not crawl private/authenticated pages without explicit approval.
  • Do not store private analytics exports, cookies, tokens, private URLs, or .env values in reports.
  • Do not claim Lighthouse, browser, crawler, or DOM evidence unless it was actually run.
  • Do not change production content, redirects, robots, sitemap, DNS, CMS, or deployment settings.

Handoff Shape

Mode:
Sources:
Evidence status:
Major findings:
SEO/LLM risks:
UX risks:
Technical constraints:
Asset inventory:
Migration risk:
Next owner:

Validation

When editing this skill, run:

python3 $CODEX_HOME/skills/.system/skill-creator/scripts/quick_validate.py $CODEX_HOME/skills/existing-site-analyzer
python3 $CODEX_HOME/skills/senior-skill-architect/scripts/lint_production_skill.py $CODEX_HOME/skills/existing-site-analyzer
python3 -m json.tool $CODEX_HOME/skills/existing-site-analyzer/evals/evals.json >/dev/null

Comments

Markdown supported