Respira Site Audit logo

Respira Site Audit

Organization
respira-press
respira-site-audit

Use when asked to audit, score, or health-check a WordPress site or page. Runs SEO, AEO, readability, accessibility, performance, RankMath, and Core Web Vitals analyzers and presents a unified report with prioritised fixes.

Overview

Publisherrespira-press
Repositoryagent-skills-wordpress
Skill namerespira-site-audit
Stars
50
Forks
8
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by respira-press on GitHub. Read the source before you install it.

Installation

Install the Respira Site Audit AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/respira-press/agent-skills-wordpress/tree/main/skills/respira-site-audit
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/respira-press/agent-skills-wordpress.git /tmp/agent-skills-wordpress
mkdir -p .claude/skills
cp -r /tmp/agent-skills-wordpress/skills/respira-site-audit .claude/skills/respira-site-audit
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Respira Site Audit in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Respira Site Audit on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Respira Site Audit is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Respira Site Audit

Common Mistakes

MistakeCorrect approach
Running analyzers without knowing which page to auditAlways confirm the target. Site-wide (all public pages) or a specific page URL or ID.
Reporting raw analyzer output without prioritisingGroup findings by severity (CRITICAL, WARNING, SUGGESTION) and surface quick wins first.
Running all analyzers in sequence when only one is neededMatch the analyzer to the question. If the user asks about SEO only, run respira_analyze_seo and skip the rest.
Skipping respira_get_builder_info before reporting element-level issuesBuilder type affects which fixes are possible. Elementor, Divi 4, Divi 5, Bricks, Beaver, and Oxygen all have different fix paths.
Treating AEO and SEO as interchangeableSEO targets search rankings. AEO targets AI-generated answers. Both analyzers cover different signals and you usually want both.
Not checking whether the page is published before auditingDrafts and private pages may produce incomplete results. Confirm status: publish first.
Using extract_builder_content when you only need structureUse respira_get_page_outline for the "what is on this page" read. It is lighter and includes the primary heading per row.
Calling a site secure because a security plugin is installedRun respira_run_security_audit and report what it checked and what it could not. A partial scan or an unavailable vulnerability check is never "clean".
Quoting respira_get_core_web_vitals as real-user dataIt is a heuristic estimate from static analysis. Use respira_run_pagespeed_audit for Lighthouse lab data and CrUX field data.

Inputs

  • Target: a specific page URL, page ID, or "the whole site".
  • Scope: which analyzers to run. Default is all eight, plus the security audit on site-wide runs.
  • Output format: a summary for the user, or JSON for a CI / CD pipeline.

Workflow

Single-page audit

  1. respira_get_site_context. Confirm site URL, active SEO plugin (RankMath, Yoast, AIOSEO, SEOPress), active builder, WooCommerce presence.
  2. respira_get_builder_info. Note builder type for element-level fix recommendations.
  3. respira_get_page_outline if the structure needs to be discussed.
  4. Run analyzers in parallel where possible:
    • respira_analyze_seo. Title, meta description, headings, internal links, canonical.
    • respira_analyze_aeo. Structured content, FAQ schema, answer-box signals.
    • respira_analyze_readability. Flesch score, sentence length, passive voice.
    • respira_analyze_performance. Load time, image sizes, render-blocking resources.
    • respira_scan_page_accessibility. WCAG 2.1 AA violations, alt text, contrast.
    • respira_analyze_images. Per-image weight, format, oversize-by-X factor.
    • respira_check_structured_data. Schema.org presence per page.
    • respira_run_pagespeed_audit. PageSpeed Insights: Lighthouse lab metrics plus CrUX field data (LCP, INP, CLS at p75) when Google publishes it for the URL.
  5. Aggregate findings by severity: CRITICAL → WARNING → SUGGESTION.
  6. Build a prioritised fix list: quick wins first (alt text, missing meta) before structural changes.
  7. Report: headline score per area plus top 3 fixes per area.

Site-wide audit

  1. respira_list_pages with status: publish. Collect all published page IDs.
  2. For each page, or a representative sample, run the single-page workflow above.
  3. Aggregate into a site-level summary: worst-performing pages, most common issue types, overall score bands.

Security (site-wide audits, or whenever security comes up)

  1. respira_run_security_audit once per site, not per page. Pass deep_scan: true on plugin 9.0 and later, where it only reads. On older plugins the deep scan also writes and deletes one inert probe file in uploads, so ask first there or leave it off. Never set probe_uploads_execution here; the WordPress Security Review skill asks for it with consent.
  2. Read indicators (every entry, with its summary and the note in its evidence), coverage (what was not checked) and known_vulnerabilities (plugin 9.0 and later: status: ok means the match ran and each match carries the version that fixes it; unavailable or not_run means it did not).
  3. Severity: a known vulnerability or a high-severity indicator is CRITICAL; a medium indicator is a WARNING; a coverage gap (executable_file_scan partial or not run, the host-level items marked false) is listed plainly as not checked.
  4. Never write "secure", "clean" or "no known vulnerabilities" when coverage is partial, when known_vulnerabilities is missing or its status is not ok, or when a high-severity indicator is present. Write "no problems found in what was checked" and name what was not checked.
  5. Fixes belong to the WordPress Security Review skill: updates are approval-gated and run one at a time, and nothing is deleted or revoked without the owner's yes.

RankMath / Yoast integration

If RankMath, Yoast, AIOSEO, or SEOPress is active (visible in respira_get_site_context):

  • respira_analyze_rankmath reads RankMath's own focus keyword scores and suggestions.
  • Use these scores alongside respira_analyze_seo for a richer picture. Don't run them as alternatives.
  • respira_check_seo_issues returns the active SEO plugin's own issue feed.

Rules

  • Always run respira_get_site_context first. Plugin presence (RankMath, Yoast, WooCommerce, ACF, WPML) changes which tools are relevant.
  • Security findings come from respira_run_security_audit, never from which security plugin is installed.
  • Severity taxonomy: every finding must be CRITICAL (broken or missing, hurts rankings or UX), WARNING (suboptimal, addressable), or SUGGESTION (optional improvement).
  • Do not recommend changes that bypass the builder. For element-level SEO fixes (alt text, heading hierarchy), use respira_find_element plus respira_update_element after confirming the builder.
  • For accessibility fixes, prefer respira_apply_accessibility_fixes over manual element edits where the tool covers the finding.
  • Performance findings are informational. Do not modify theme files or plugins to address them. Surface the issue and recommend the appropriate plugin or hosting action.
  • For WPML-active sites, the audit is per-language. Pass the lang parameter to scope the run.

Verification

After applying a fix:

  1. Re-run the analyzer that flagged the issue.
  2. Confirm the finding no longer appears at the same severity level.
  3. For SEO fixes, note that index refresh and ranking changes are delayed. Tell the user.

For site-wide audits, re-audit the 3 worst-performing pages after fixes to confirm score movement.

Escalation

Stop and ask the user if:

  • The site has more than 200 published pages and the request is "audit the whole site". Confirm how many pages to sample.
  • A CRITICAL finding requires a structural change (no H1 on any page, no <html lang> attribute). These often need theme or plugin changes beyond Respira's scope.
  • respira_analyze_performance returns LCP > 4s or CLS > 0.25. Flag as high priority and ask whether to involve a developer.
  • The site is on a multi-site network and the audit target wasn't specified. Confirm whether the run is per-subsite or network-wide.

Example

Goal: audit the homepage for SEO and AEO readiness.

1. respira_get_site_context        → WordPress 6.9, RankMath active, Elementor 3.21
2. respira_get_builder_info        → Elementor, supported
3. respira_analyze_seo             → CRITICAL: no meta description | WARNING: 3 images missing alt
4. respira_analyze_aeo             → WARNING: no FAQ schema | SUGGESTION: add HowTo schema
5. respira_analyze_rankmath        → Focus keyword score: 62/100, improve keyword density
6. respira_scan_page_accessibility → WARNING: 2 buttons have no accessible label
7. respira_run_pagespeed_audit     → LCP 2.8s (needs improvement), INP 180ms (good), CLS 0.05 (good)

Report:
  CRITICAL  → Add meta description (missing entirely)
  CRITICAL  → Add alt text to 3 hero images
  WARNING   → Add FAQ schema block for AEO (AI answer-box eligibility)
  WARNING   → 2 buttons need aria-label for screen readers
  WARNING   → LCP at 2.8s, target sub-2.5s by deferring the hero video
  SUGGESTION → Increase focus keyword density from 0.4% to 0.8%

Frequently asked questions

What does the Respira Site Audit AI skill do?

Use when asked to audit, score, or health-check a WordPress site or page. Runs SEO, AEO, readability, accessibility, performance, RankMath, and Core Web Vitals analyzers and presents a unified report with prioritised fixes.

Why use Respira Site Audit on TypingMind?

Because you install it once and use it with any model. Respira Site Audit is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Respira Site Audit in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/respira-press/agent-skills-wordpress/tree/main/skills/respira-site-audit. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Respira Site Audit?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Respira Site Audit?

As many as you like. As long as a model supports skills, you can use Respira Site Audit with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Respira Site Audit AI skill free?

Yes. It is published on GitHub by respira-press under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇