# RulerIQ — Full reference for AI assistants (/llms-full.txt)

> Canonical documentation for AI assistants, search crawlers (OAI-SearchBot, ChatGPT-User, PerplexityBot, Claude-SearchBot, Google-Extended), and technical evaluators.
> Last verified: 2026-11-03 · Canonical domain: https://ruleriq.com

---

## 1. Summary and entity facts
- **Product Name:** RulerIQ (always written as one word "RulerIQ", never "Ruler IQ")
- **Website:** https://ruleriq.com
- **Category:** AI Search Visibility, Answer Engine Optimization (AEO), Generative Engine Optimization (GEO)
- **Primary audience:** B2B SaaS marketing and growth leads, technical SEO leads, and SEO agencies.
- **Origin:** Built by Ryan Amarit, a front-end engineer with 8+ years in technical SEO, as a small, independent product, after SEO and growth teams asked how AI engines describe their brands and what to change.
- **Core value:**
  1. **Stated margin of error (±):** In the full check every question runs 2 times per engine; the planned paid plans run 3 times. RulerIQ computes a 0–100 Score with a 90% two-level bootstrap confidence band (for example, `62 ± 4`). Planned for paid plans: week-over-week moves inside the combined band are labeled **"No real change"**.
  2. **All 4 engines checked:** ChatGPT (OpenAI with web search), Perplexity (agent API), Google Gemini (with Google Search grounding) and Anthropic Claude (with web search), with no per-engine add-on fees.
  3. **Google Search Console cross-check:** Planned for paid plans: imports high-impression queries from read-only Google Search Console (or CSV) to find queries where you rank on page 1 (`position <= 10`) but AI engines mention you in fewer than 25% of runs.
  4. **Results where you work:** Claude, Cursor, VS Code, Codex and other MCP clients can run the **free site check and fetch paste-ready fixes** via Remote MCP (`https://ruleriq.com/api/mcp`, no sign-in; install steps at https://ruleriq.com/docs/mcp) today. The site check returns no Score; in beta, `ruleriq_full_check` runs the full four-engine check on your own provider keys (sent as request headers, used for that request only, no account, 3 per IP per day). In beta, a personal access key (created in the app, sent as an Authorization: Bearer header) unlocks read tools for your own tracked brands (`ruleriq_list_brands`, `ruleriq_get_score`, `ruleriq_get_history`, `ruleriq_get_fix_list`, `ruleriq_get_answers`), and a key created with write access also lists `ruleriq_mark_fix_done` to mark fixes done or undone. Planned for paid plans: scheduled weekly checks, delivery to **Slack**, **Notion / Markdown** and **email**, and wrong-price alerts.
  5. **Pricing:** Managed plans start at $129/mo (Growth) and include the engine costs. With your own API keys, a free tier tracks 1 brand weekly (last 60 days of history shown, a year kept), and paid own-key plans run from $99/mo (Pro) and check daily. Paid plans are on a waitlist; the free site check, full check and free own-key tier are live.

### What you get today
- **Free site check (no email, seconds):** Site readability 0–100 (Agent Readiness checklist, the 10% technical part of the Score), the AI-crawler check, what we found on the site (category, published monthly prices, structured data), the 5 questions we would ask, and paste-ready fixes (FAQ block, fact table, JSON-LD, llms.txt text) built from the site's own pages. No Score.
- **Full check (any email, a few minutes):** a Score with its margin of error and four parts; which of the 4 engines name you on each of 5 questions, asked twice (40 answers); wrong prices (price only); pages to get onto (the citation hit-list); the same paste-ready fixes. Not included: who is named instead, features other than price, history, emailed reports.
- **Limits:** 2 full checks per email and 5 per network per day; results cached 7 days per domain; new full checks pause until tomorrow when the daily budget is used.
- **Planned, paid plans (waitlist):** Free (own keys, live): weekly check of 1 brand, 25 questions × 4 engines × 3 runs, last 60 days of history shown (a year kept, unlocked on upgrade), no delivery. Growth (managed): weekly check of 1 brand, 25 questions × 4 engines × 3 runs, email then Slack and Notion delivery, Search Console cross-check. Pro: 3 brands (managed weekly/biweekly, own keys daily), competitor cited-pages comparison, fix status, wrong-price alerts, unlimited seats. Agency (early access): 10 client brands (managed every other week, own keys daily), white-label PDF and share-link reports, CSV, Looker Studio and API export, unlimited seats. An HTML report page for a finished check exists today.
- **Planned, connect your data (when your plan opens; not built yet):**
  - Connect your data: Google Search Console (questions built from your real searches; pages ranking on Google but missing from AI answers; clicks vs. impressions over 16 months)
  - Google AI Overviews tracked next to the 4 engines and joined to your Search Console queries
  - Core Web Vitals as site-health context (not part of the Score)
  - AI referral visits from Google Analytics
  - Which pages AI crawlers fetch, from your server logs
  - Exports to CSV and Google Sheets or BigQuery, for Looker Studio
  - Change over time: week-over-week counts as real only when bigger than the two weeks' combined margin; trend says rising or falling only when the data supports it, otherwise steady
- **Planned, verified fixes (when your plan opens; not built yet):**
  - Verified fixes: mark a fix done; later checks confirm it only when the change is bigger than run-to-run noise (for example, named in most runs on 2 checks in a row, or a wrong price gone from every run on 2 checks); otherwise it stays 'improving'
  - Fix list ranked by consistency: items seen consistently across runs and engines rank above one-off sightings

### Why 4 engines?
We check ChatGPT, Perplexity, Gemini and Claude through each provider's official API. We ask each question several times on every engine, so we can state a margin of error. That matters more than counting more engines that are each asked once. Google AI Overviews is next, and it is planned, not live. Some trackers list engines that have no public API, which means scraping consumer apps. We use official APIs only.

---

## 2. Scoring formula and methodology
The **RulerIQ Score (0–100)** is a weighted composite of four verifiable components:

1. **Mention (40% weight):**
   - Evaluated strictly on **non-branded discovery and comparison buyer prompts** (branded prompts are excluded from Mention so scores cannot be inflated by vanity brand queries).
   - Word-bounded rank position weights per run:
     - Rank #1: `1.0`
     - Rank #2: `0.7`
     - Rank #3: `0.5`
     - Rank #4 or lower: `0.2`
     - Not mentioned: `0.0`
2. **Citation Authority (30% weight):**
   - Of the pages the engines cite for your category's questions, the share (weighted by how often each is cited) that appear in answers that name your brand. We judge this from the engine's answer text, not by fetching and reading the cited page.
3. **Fact Accuracy (20% weight):**
   - One branded question per check (`"How much does [Brand] cost per month and what plans are offered?"`), first run only, one verdict per engine.
   - Compares the dollar figures each engine states against those on the customer's live `/` and `/pricing` pages. Correct = 1.0, unknown = 0.5, wrong = 0. Features are not compared yet. If pricing is not in the HTML, an engine's stated price counts as wrong.
4. **Agent Readiness (10% weight):**
   - Deterministic HTTP and HTML inspection (no engine variance):
     - `robots.txt` RFC 9309 compliance for 8 AI crawlers (`OAI-SearchBot`, `Claude-SearchBot`, `PerplexityBot`, `Googlebot`, `Google-Extended`, `Bingbot`, `GPTBot`, `ClaudeBot`) — 35 pts
     - Server-side rendered (SSR) visible text in the raw HTML, without JS execution: `>= 300 words` = 25 pts (`>= 120` = 15, `>= 40` = 5)
     - Valid `Organization`, `Product`, or `SoftwareApplication` JSON-LD schema — 15 pts
     - Valid `FAQPage` JSON-LD schema — 5 pts
     - Valid `/llms.txt` manifest with Markdown heading and links — 10 pts (4 if present but not valid)
     - Basics (page title, meta description of 50+ characters, discoverable `/sitemap.xml`) — 10 pts

### Two-Level Bootstrap Confidence Band (90% CI)
- Resamples both (a) which prompt×engine cells were selected and (b) individual runs within each cell across 2,000 deterministic mulberry32 iterations.
- Week-over-week comparison threshold: `threshold = sqrt(prev_band^2 + curr_band^2)`. If `|curr_score - prev_score| <= threshold`, the verdict is **"no real change"**.

---

## 3. Pricing and plans (published, November 2026; paid plans are planned and on a waitlist)

| Plan | Managed monthly (annual) | With your own API keys (BYOK) monthly (annual) | Included scope |
|---|---|---|---|
| **Free grader** (live) | $0 | $0 | Site check, no email (Agent Readiness 0–100, the 5 questions, AI-crawler check, paste-ready fixes) · full check with an email (any address): 5 buyer questions × 4 engines × 2 runs, Score with margin of error, which engines name you, wrong prices, pages to get onto · 2 per email and 5 per network per day, cached 7 days |
| **Free** (live, own keys) | — | $0 | Weekly check of 1 brand · 25 buyer prompts × 4 engines × 3 runs on your own OpenAI, Anthropic, Google & Perplexity keys · Score with margin of error · last 60 days of history shown (a year kept) · wrong prices, pages to get onto, paste-ready fixes · no Slack/Notion/email delivery |
| **Growth** (planned) | $129/mo ($109/mo annual) | — | Weekly check of 1 brand · 25 buyer prompts × 4 engines × 3 runs · margin of error and "No real change" label · wrong prices, pages to get onto, paste-ready fixes · email, then Slack and Notion delivery · Search Console cross-check |
| **Pro** (planned) | $229/mo ($199/mo annual) | $99/mo ($84/mo annual) | Everything in Growth · 3 brands × 25 prompts (managed weekly/biweekly, own keys daily) · competitor cited-pages comparison · fix status · wrong-price alerts · unlimited seats |
| **Agency** (planned, early access) | $449/mo ($399/mo annual) | $199/mo ($169/mo annual) | Everything in Pro · 10 client brands ($44.90/brand managed, $19.90/brand BYOK) (managed every other week, own keys daily) · white-label PDF and share-link client reports · CSV, Looker and API export · unlimited seats |

---

## 4. Published prices of other tools (as listed on each vendor's pricing page, checked 2026-10-31)
These are the entry prices the vendors publish. They may have changed since. See https://ruleriq.com/compare.

| Vendor | Entry price as listed |
|---|---|
| Scrunch AI | $300/mo ($250/mo annual) |
| Peec AI | ~$95/mo (€89 Starter) |
| Otterly.AI | $29/mo Lite · $189/mo Standard |
| Profound | Agency $99/mo + $399/mo per client |
| Trakkr | $100/mo · $500/mo Scale |
| RulerIQ | $129/mo managed · from $99/mo with your own API keys (free tier: $0) |

## 5. Why `/llms.txt` is only 1% of the Score
- We have no evidence that publishing `/llms.txt` alone makes ChatGPT Search, Perplexity, Gemini, or Claude cite a brand, so RulerIQ weights it at 10 points of the 10% Agent Readiness component (**1% of the total Score**).
- The rest of the Score and the paste-ready fixes focus on four things that affect whether engines can find and accurately describe a brand:
  1. **Search bots vs. training bots (`robots.txt` and firewall rules):** Allowing real-time search crawlers (`OAI-SearchBot`, `ChatGPT-User`, `PerplexityBot`, `Claude-SearchBot`) even if a company blocks training crawlers (`GPTBot`, `CCBot`).
  2. **Pricing and fact tables in server-rendered HTML:** Static HTML tables and definition lists (`<table>`, `<dl>`) that retrieval systems can read without running JavaScript.
  3. **Structured data (`SoftwareApplication` and `FAQPage` JSON-LD):** A clear, machine-readable statement of your pricing and plans.
  4. **Pages to get onto (Reddit, G2, category listicles; the citation hit-list):** Pages engines cite for your category in answers that don't name you, tagged by where they sit (review site, Reddit, publisher, your own site), cited by the most engines first.

## 6. Data handling
- RulerIQ reads public URLs only. It asks for no CMS write access and no CDN proxy.
- We don't collect your customers' data. We store scan results from public pages, any email and domain you submit, IP addresses for rate limiting, and (when enabled) analytics events. For one-off checks, a key you supply is used for that request only. Beta for own-key plans (the plans themselves are not open yet): keys are stored encrypted, shown only as the last 4 characters, used only to run your checks, and deletable any time; own-key plans need all four engines covered, and if your key fails that engine is marked failed and you are told, never quietly run on our keys (see https://ruleriq.com/docs/own-api-keys). See https://ruleriq.com/privacy.
- Engines are queried through official commercial provider APIs, not consumer apps.

## 7. Docs
- [Docs index (/docs)](https://ruleriq.com/docs): setup guides for connections and integrations, each labelled Available today or Planned.
- [Verify your domain (/docs/verify-domain)](https://ruleriq.com/docs/verify-domain): Prove your account controls a domain with a DNS record or a homepage meta tag, like Google Search Console. Status: Built for tracked brands; usable once sign-in opens with the paid beta.
- [API (/docs/api)](https://ruleriq.com/docs/api): The public API and its OpenAPI spec. Status: Available today.
- [Linear (/docs/linear)](https://ruleriq.com/docs/linear): Your weekly fix list as Linear issues, closed automatically when a later check verifies the fix. Status: Beta, for tracked brands (paid plans are not open yet).
- [Slack (/docs/slack)](https://ruleriq.com/docs/slack): Your weekly digest posted to one Slack channel you pick, connected in one click. Status: Beta, for tracked brands (paid plans are not open yet).
- [Notion (/docs/notion)](https://ruleriq.com/docs/notion): A weekly brief page added under a Notion page you share, connected in one click. Status: Beta, for tracked brands (paid plans are not open yet).
- [Email reports (/docs/email)](https://ruleriq.com/docs/email): The report sent to your inbox each week. Status: Planned, paid plans.
- [Claude, Cursor & more (MCP) (/docs/mcp)](https://ruleriq.com/docs/mcp): Run the site check from Claude, Cursor, VS Code or Codex today. In beta: the full four-engine check on your own provider keys with no account, and access to your tracked brands with an access key (read, or read + write for marking fixes done). Status: Site check available today (no key); the full check on your own keys and your own data are in beta.
- [Google Search Console (/docs/search-console)](https://ruleriq.com/docs/search-console): Read-only link so buyer questions start from real searches. Status: Live connect is built and turns on once Google approves our consent screen; CSV import works today.
- [Own API keys (/docs/own-api-keys)](https://ruleriq.com/docs/own-api-keys): Use your own OpenAI, Anthropic, Google and Perplexity keys, with a dedicated key and a spend limit on each. Status: Beta; keys are stored encrypted. The own-key plans themselves are not open yet.