🧰 Page Meta Check
Checks the title, meta description, canonical, robots, H1 and Open Graph tags of one page and returns a pass/fail table with the top fixes.
What it does
Give your agent a URL, a local HTML file or pasted page source. The skill reads the page head and the first H1, checks seven items against clear thresholds and returns a table with the value it found for each one. It ends with up to five fixes, most important first.
- Works on one page per run.
- Reads only. It does not change your files.
- Flags
noindexwithout assuming it is a mistake, because staging and private pages often use it on purpose.
Output preview
Task given to the agent: “Can you check the SEO meta tags on index.html in this folder and tell me what to fix? Just report, do not edit the file.”
The biggest issue is **`noindex`**, which prevents this page from appearing in search results. No files were edited. | Priority | Location | What to fix | |---|---|---| | Critical | [index.html:6](index.html:6) | Remove `noindex` if this is a public page intended for search. `follow` does not override it. Keep it if exclusion is intentional. | | High | [index.html:7](index.html:7) | Verify that `/shop/` is the intended canonical page. Use its full absolute HTTPS URL. If this is a distinct homepage, its canonical should normally point to its own public URL. | | Medium | `<head>` | Add a unique meta description, such as: “Discover handmade ceramic mugs, bowls and plates from Clay & Kiln Studio. Small-batch stoneware made for everyday use.” | | Medium | `<head>` | Add `<meta name="viewport" content="width=device-width, initial-scale=1">` for proper mobile sizing. | | Optional | [index.html:8](index.html:8) | Make `og:title` describe the page, and complete the social preview with `og:description`, `og:image`, `og:url`, and `og:type`. | The existing title is descriptive and includes the brand. Shortening it could improve readability, but it is not an urgent fix. Google documents the effects of [robots directives](https://developers.google.com/search/docs/crawling-indexing/robots-meta-tag), [canonical URLs](https://developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls), and [description and viewport tags](https://developers.google.com/search/docs/crawling-indexing/special-tags).
Page: [index.html](index.html). No files changed. | Check | Result | Found | Fix | |---|---|---|---| | Title | FAIL | “Handmade Ceramic Mugs, Bowls and Plates for Everyday Use \| Clay & Kiln Studio” — 77 characters | Shorten to 30–60 characters. | | Meta description | FAIL | Missing | Add a relevant description of 70–160 characters. | | Canonical | WARN | `/shop/` | Use the absolute URL of the intended canonical page. | | Robots | FAIL | `noindex, follow` | **Blocks indexing.** Confirm whether this is intentional before changing it. | | H1 | WARN | 2 headings; first: “Handmade ceramics” | Keep one main H1. | | HTML language | FAIL | Missing | Add `lang="en"` if the page is in English. | | Open Graph | WARN | `og:title="Clay & Kiln Studio"`; description and image missing | Add `og:description` and `og:image`. | Top fixes: 1. Confirm whether the page should appear in search results before changing `noindex`. 2. Shorten the title and add a meta description. 3. Replace `/shop/` with the correct absolute canonical URL. 4. Add the HTML language and keep one main H1. 5. Complete the Open Graph tags for shared-link previews.
Real, unedited output from recorded runs on Oct 18, 2026 · Codex (codex-cli 0.163.0) · model gpt-6.1-sol. Test inputs are fixtures (fictional brand, example.com links). AI output can be wrong: check facts before you publish anything like it.
Install
Unzip first. You get a folder named page-meta-check containing SKILL.md.
Claude Code
Put the folder in your personal skills directory, or in the project to share it with your team.
~/.claude/skills/page-meta-check/SKILL.md.claude/skills/page-meta-check/SKILL.mdClaude Code picks up new skills in the current session (run /reload-skills if the skills folder didn't exist when the session started). The skill loads when a task matches its description.
Claude.ai
Turn on code execution (Settings → Capabilities), then go to Customize → Skills, click +, then Create skill → Upload a skill, and select the ZIP. Full guide.
Cursor
Put the folder in your user or project skills directory. Cursor also reads .claude/skills.
~/.cursor/skills/page-meta-check/SKILL.md.cursor/skills/page-meta-check/SKILL.mdCodex
Put the folder in your user or repository skills directory, then restart Codex or run /skills to check it is listed.
~/.agents/skills/page-meta-check/SKILL.md.agents/skills/page-meta-check/SKILL.mdThese are the standard skill locations for each tool. The tools a skill is verified on are listed under “Tested on” above, if any.
Test log
| Date (UTC) | Tool | Model | Run | Result | Checks | Tokens in / out |
|---|---|---|---|---|---|---|
| Oct 18, 2026 | Codex | gpt-5.4 | Test | Failed | 2/12
| 24,360 / 617 |
| Oct 18, 2026 | Codex | gpt-5.4-mini | Test | Failed | 2/12
| 36,682 / 465 |
| Oct 18, 2026 | Codex | gpt-5.4-mini | Test | Failed | 2/12
| 36,386 / 423 |
| Oct 18, 2026 | Codex | gpt-6.1-sol | Baseline (no skill) | Comparison only | 2/11
| 59,056 / 598 |
| Oct 18, 2026 | Codex | gpt-6.1-sol | Test | Passed | 12/12
| 57,572 / 968 |
Each row is a real run recorded by our test runner. For each tool, the latest test run decides whether it appears under “Tested on”. Baseline runs do the same task with no skill installed and never count.