AI Search Readiness Check

Source: a small local plumbing business site.

Before: "Can ChatGPT and other AI search tools actually read my client's site?"

After (skill output, unedited, excerpt): "## AI search readiness: 85/100 [...] 1. Add Organization or LocalBusiness structured data (15 points lost). The home page's JSON-LD only declares WebSite. LocalBusiness (or the more specific Plumber type) states who the business is in a form machines can read. It must use [the business's] real name, address and phone number."

Free from Pro Skill Packs, for marketers and small agencies who fix client websites. The Page Fix Sheet (9 USD) checks one page's title, description, headings, canonical and structured data, and writes replacements from the page's own facts:

Give it a website address. It checks whether AI search tools such as ChatGPT search, Perplexity, Claude and Google can read the home page, and gives a score out of 100 from 8 checks, the evidence for each, and the fixes in order. A script does the fetching and scoring, so the same site gets the same score.

Tested on 8 real public websites, as a skill and (four of them) as a pasted prompt, two also on a second model, then re-scored after two independent reviews. The score matched our script in every run. Final scores: 8.5, 8.5, 8.5, 8.5, 9, 9, 9 and 9 out of 10.

8 real tests on public websites: a small trade site, a dental practice, two small software companies, a bakery, a US vet hospital, a US dentist, and a US dental practice whose robots.txt keeps automated readers out. Each was run as a skill, four also as the pasted prompt version and two on a second model. The score and every point matched the script in every run. Final scores 8.5, 8.5, 8.5, 8.5, 9, 9, 9 and 9 out of 10 (prompt version: 9, 8.5, 9 and 8.5), under a strict rubric where one wrong fact caps a test at 6. An independent review found the prompt version stated a question count it had guessed; counts made by reading are now marked as estimates and kept out of the client summary. A second review found the script missed structured data typed as a specific business (a vet, a dentist); it now uses schema.org's full list of business types.

The tests ran on real public data and were scored against a written rubric.

How we test, what we do not do, and which assistants we ran it on.

• It checks the home page and robots.txt only, and can't test firewalls or bot protection.
• The score is our checklist with our own weights, not a score from any AI company, and promises no citations or traffic.

Works with any assistant that reads SKILL.md skills (Claude, Codex, Gemini CLI, Cursor and others), plus a paste-in prompt version for ChatGPT or any chatbot. We run our release tests on Claude Sonnet and Opus. Use your assistant's strongest model: in our tests, a small, fast model added facts that were not in the source data. The skill needs Python 3 (standard library only) and web access. The paste-in prompt needs neither.

1. Run npx skills add proskillpacks/skills in your project and pick this skill. Or copy the ai-search-readiness-check folder from GitHub (https://github.com/proskillpacks/skills/tree/main/skills/marketing-seo/ai-search-readiness-check).
2. Put the ai-search-readiness-check folder where your assistant looks for skills. The generic path is .agents/skills/ (or npx skills add from a folder). Tool-specific paths:
   • Claude Code: ~/.claude/skills/ (or .claude/skills/ inside one project)
   • OpenAI Codex, Gemini CLI, Cursor and others: .agents/skills/ or the skills folder their docs name
   • claude.ai: zip the skill folder yourself, then Customize, Skills, the plus button, Create skill, Upload a skill (code execution must be on in Settings, Capabilities)
   • ChatGPT or any chatbot: open the skill's prompt.md, paste it into the chat, then paste your own text or link
   • Steps for each tool, with links to its official docs: how to install Agent Skills
3. Ask for the job in plain words, for example: "Check whether AI search tools can read https://example.com and tell me what to fix.". The skill loads by itself.