1-Click Connect, Autopilot & MCP

WhyIQ runs two products that diagnose why you lose buyers and AI search visibility:

WhyIQ CRO Scanner

Captures your page in a real headless browser, simulates 50 visitors across four buyer archetypes, and predicts on-page AI readiness via the AI Citability Index.

WhyIQ AI Radar

Measures your weekly AI visibility (the share of AI answers that named or linked to you) across ChatGPT, Perplexity, Google AI Overviews and Google AI Mode, Gemini, and Claude.

1-Click Connect, the 4-Mode Autopilot Switchboard, and the WhyIQ MCP Server close the execution loop for both products. Instead of copying findings into tickets by hand, you can connect the platform where your site lives, choose which categories of fixes can deploy automatically versus stage as drafts for your review, or let your coding agent pull ranked evidence and ship verified fixes directly inside your IDE.

When you run a free page scan, WhyIQ gives you two ways to unlock the full 50-visitor report on your first domain: pay the one-time $9 First Scan promo, or verify domain ownership in one click through 1-Click Connect.

Connecting and verifying your live site unlocks the Full 50-Visitor CRO Report plus your 1st Automated Fix Free on any deterministic technical fix (such as JSON-LD structured data, /llms.txt discovery files, canonicals, or image alt text) that costs $0 in LLM generation.

How ownership verification prevents abuse

1 domain per accountFree

Each authenticated account can claim one Connected Free Tier domain. Additional client domains use a paid Scanner or AI Radar plan.

1 domain globally across accounts

Once a root domain has claimed its Connected Free Tier unlock, creating a second free account for the same domain will not grant another free unlock.

Verified write or DNS/meta ownership

Your connected GitHub repo, CMS token, Cloudflare route, or MCP verification token must match the scanned domain.

Live production domain check

Empty template repos or newly created sandboxes that do not serve the scanned domain's live HTML do not qualify until connected to the real site.

Deterministic first fix ($0 LLM cost)

Your complimentary automated fix runs on our deterministic AST and schema generators (JSON-LD, llms.txt, meta/canonical tags, alt text). Full AI copywriting and multi-page generation require a paid plan.

Most founders and marketers do not want to copy JSON configs or burn AI builder credits pasting long prompts. The 1-Click Connect wizard detects your stack and gives you a native connection path for all five ways modern sites are built:

Lovable, Bolt & v0 (2-Way GitHub Sync · 0 builder credits used)

Lovable, Bolt, and v0 feature native two-way GitHub sync. When you connect your project's GitHub repo in 1 click, WhyIQ commits JSON-LD schema, meta tags, and copy fixes directly to your repo. Lovable syncs those commits into your live builder automatically without spending a single Lovable message credit.

Replit Agent & Repls (Git Sync or 1-Click Agent Prompt · 0 checkpoints used)

Connect your Repl's linked GitHub repository so WhyIQ pushes clean commits that your Repl pulls automatically without burning paid Replit Agent checkpoints, or copy a pre-scoped single-turn Replit Agent prompt.

Webflow, WordPress, Shopify & Cloudflare Edge (Safe Draft Mode)

Connect Webflow, WordPress, or Shopify via OAuth to stage copy, FAQ, and meta updates in Safe Draft Mode so nothing goes live until you click Publish in your CMS. Pair Cloudflare Workers at the edge to inject JSON-LD, /llms.txt, 301 redirects, and pre-rendered HTML for AI search crawlers (GPTBot, PerplexityBot, ClaudeBot) without touching your CMS theme.

GitHub / Next.js / Astro / Remix (Syntax-Checked Pull Requests)

Install the WhyIQ GitHub App on your repository. WhyIQ opens atomic, syntax-checked Pull Requests (or commits directly for deterministic fixes you set to Auto-deploy) that trigger your existing Vercel or Netlify preview deployments.

Claude, ChatGPT, Cursor & Windsurf (1-Click MCP Connector)

Click Connect to Cursor, Connect to Claude Desktop / Claude Code, or copy the Streamable HTTP URL (https://www.whyiq.ai/api/mcp) into ChatGPT Developer Mode so your AI assistant can inspect your scores and apply guarded fixes locally.

Not every fix carries the same risk. Adding Organization JSON-LD or /llms.txt in the <head> does not change a single pixel of your visible design, whereas rewriting your hero headline or publishing a /vs/competitor page should wait for your sign-off unless you explicitly ask for hands-free execution.

The 4-Mode Autopilot Switchboard lets you set the exact autonomy level for 4 Product Groups and 18 granular fix elements, or apply one of three presets in a single click.

The four autonomy modes

Off

WhyIQ does not generate or queue fixes for this element.

Flag only

WhyIQ diagnoses the issue in your CRO report or AI Radar Fixes list and shows the evidence, but does not draft code or CMS changes.

Stage for review

WhyIQ generates the exact code diff, Pull Request, or CMS Safe Draft and holds it in your approval queue until you click Approve.

Auto-deploy

WhyIQ snapshots the pre-change SHA-256 hash, syntax-checks the patch, deploys it directly (when it passes the Honesty Gate), and verifies it on the next scan.

The 4 Product Groups & 18 controllable elements

Group 1 · Technical & Schema Foundations (Default: Auto-deploy)

Includes (1) JSON-LD Structured Data (Organization, Product, SoftwareApplication, FAQPage, Article), (2) llms.txt & AI Discovery Endpoints (/llms.txt and /llms-full.txt), (3) Meta Titles, Descriptions, OpenGraph & Canonicals, (4) Image Alt Text, ARIA Labels & Tap-Target Accessibility Fixes, (5) 301 Redirects & Broken Internal Link Repair, and (6) Edge Pre-Rendering for AI Search Crawlers (GPTBot, PerplexityBot, ClaudeBot).

Group 2 · On-Page Copy, CRO & Answer Blocks (Default: Stage for review)

Includes (7) Hero Headline & Value Proposition (5-second clarity), (8) Primary CTA Button Copy, Placement & Microcopy, (9) Incoming Ad / AI Citation Scent-Match Alignment, (10) Direct 40–60 Word Answer Blocks & FAQ Sections for Lost AI Prompts, (11) Trust Signals, Proof Placement & Risk-Reversal Framing, and (12) Pricing & Packaging Table Clarity.

Group 3 · New Pages & Programmatic Content (Default: Stage for review)

Includes (13) Competitor Comparison & Alternative Pages (/vs/* and /alternatives/*), (14) Dedicated Buyer-Question Landing Pages (when AI Radar Your site flags 'Needs a new page'), (15) Glossary & Entity Definition Pages, and (16) Case Study & Proof Page Scaffolding.

Group 4 · Off-Site Authority & AI Radar Actions (Capped at Stage for review)

Includes (17) Reddit & Community Thread Replies (drafted for threads AI engines actually cited) and (18) Publication Pitches, G2/Directory Listings & Prompt Bank Updates. Off-site community and PR actions are permanently capped at Stage for review so WhyIQ never posts under your brand name without human sign-off.

Every automated or agent-assisted fix passes through three non-negotiable safety rails enforced on the server before a single byte touches your site:

1. 1

   The Honesty Gate (zero fabricated proof)

   WhyIQ never invents customer testimonials, client logos, star ratings, SOC 2 / HIPAA certifications, SLA numbers, or pricing tiers. Whenever a recommendation requires a real business fact not already verified on your domain, WhyIQ inserts an explicit [fill in: ...] placeholder and automatically downgrades the fix from Auto-deploy to Stage for review until you supply the real text.

2. 2

   Pre-deployment SHA-256 snapshot & syntax check

   Before applying any change (via GitHub, CMS API, Cloudflare Edge, or MCP), WhyIQ records the target file path, the pre-change content, and its SHA-256 hash in your domain's Rollback Ledger, and validates JSON-LD and JSX/HTML syntax so malformed markup cannot ship.

3. 3

   1-click rollback & post-fix verification

   Every deployed fix appears in the Rollback Ledger inside your Autopilot drawer (and via the whyiq_rollback_fix MCP tool). One click restores the exact pre-change file snapshot. When your next CRO scan or weekly AI Radar run completes, WhyIQ verifies whether the fix resolved the finding and tracks your 14-day before/after conversion lift when GA4 is connected.

If you build in Cursor, Claude Code, Claude Desktop, Windsurf, or ChatGPT Developer Mode, you can connect directly to WhyIQ's remote Streamable HTTP Model Context Protocol (MCP) server at https://www.whyiq.ai/api/mcp (published manifest at /mcp/server.json).

Unlike raw data-dump MCP servers that flood your coding agent's context window with 50,000 tokens of JSON, WhyIQ uses a 3-tier progressive-disclosure architecture:

Quick configuration for Cursor, Claude & Windsurf

Click the 1-Click Connect to Cursor button in your Autopilot drawer, or add this block to .cursor/mcp.json or claude_desktop_config.json:

{
  "mcpServers": {
    "whyiq": {
      "url": "https://www.whyiq.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_WHYIQ_TOKEN>"
      }
    }
  }
}

All 8 MCP tools

1. whyiq_get_site_overview (Tier 1 · Read-only)

Returns the unified executive summary for a domain: 4-pillar CRO Scanner scores (WhyIQ Score, AI Citability Index, Search Ranking, Accessibility Score), CTA Pull (/ 100), AI Radar weekly AI visibility (Overall, Non-brand, Brand), #1 competing domain, and open Autopilot fix counts.

2. whyiq_inspect_radar_intelligence (Tier 2 · Read-only)

Queries AI Radar by view: 'lost_prompts' (questions where rivals beat you), 'competitor_diff' (what the winning rival page had that yours lacked), 'citations' (third-party domains AI engines cited), 'site_coverage' (Has a page / Needs a section / Needs a new page), or 'post_click_check' (Scent gap / Needs CTA / Fetch error on cited URLs).

3. whyiq_inspect_visitor_simulation (Tier 2 · Read-only)

Queries the 50-visitor CRO simulation by lens: 'archetypes' (per-archetype bounce rate, verbatim quotes, Believed vs. Questioned, and Win plays), 'objections' (unaddressed buyer objections and scroll-depth drop-off), 'scent_match' (incoming ad/prompt promise match), 'agent_readiness' (AI browser agent task completion), or 'pillar_breakdown' (8 AI Citability dimensions, SEO, and WCAG issues).

4. whyiq_list_autopilot_fixes (Tier 2 · Read-only)

Lists ranked fixes across CRO Scanner and AI Radar filtered by Product Group, Switchboard mode (auto_deploy, stage_review, flag_only), or deterministic ($0 LLM cost) status.

5. whyiq_prepare_fix_bundle (Tier 3 · Guarded write prep)

Prepares a specific fix ID for execution: returns the target element/URL, exact evidence, Switchboard permission, Honesty Gate inspection (flagging any [fill in: ...] placeholders), and coding instructions.

6. whyiq_verify_and_record_fix (Tier 3 · Verification & Snapshot)

Records the pre-change file content and SHA-256 hash into the Rollback Ledger, validates the applied diff against the Honesty Gate, and marks the fix ready for next-scan verification.

7. whyiq_rollback_fix (Tier 3 · Instant revert)

Looks up any ledger entry ID and returns or restores the exact pre-deployment file content and SHA-256 hash.

8. whyiq_configure_switchboard (Tier 3 · Configuration)

Applies a Switchboard preset ('recommended_safe', 'hands_free', 'review_all') or updates individual element modes.

Ambient @whyiq resources & slash prompts

whyiq://rules/guardrails.mdc

Drop-in Cursor/Claude rulebook enforcing 5-second hero clarity, CTA Pull standards, AEO answer-first formatting, and the Honesty Gate on every commit.

whyiq://context/radar-scorecard.json & whyiq://context/cro-simulation.json

Live JSON snapshots of your weekly AI Radar visibility scorecard and 50-visitor CRO simulation for @-mentioning in chat.

whyiq://context/autopilot-queue.json

The prioritized queue of open CRO and AI Radar fixes with their current Switchboard modes.

/whyiq:sync · /whyiq:why-losing · /whyiq:precommit-check

Three built-in MCP slash prompts: pull and apply today's top safe fixes (/whyiq:sync), explain why a rival is beating you on a specific buyer question (/whyiq:why-losing), or audit your staged git diff for CRO/AEO regressions before pushing (/whyiq:precommit-check).