AIO.GEO measures how artificial intelligence search engines parse, understand, and cite your digital entity. By evaluating structural DOM semantics, JSON-LD schema graphs, and crawler access policies, our protocol generates dry-run remediation fix packs sealed with cryptographic proof receipts to establish machine authority.
Know what LLMs see. Fix it. Prove it.
Start here · 10 seconds
npx @aio-geo/cli@0.3.4 audit yourdomain.comAgencies: multi-client control room · ChatGPT auditor · First HTML · mystery shopper
10-second audit
See where your site is ready for AI answers: and where it is invisible.
Competitor teardown
See why AI quoted them and not you, on one page.
Fix and prove lift
Preview concrete fixes, rescore, and verify results with a sealed receipt.
Agency control room
Run client rosters, send white-label weekly digests, and deliver $5k-$15k/mo retainers.
Select your operational lane to deploy specialized AI search readiness tools. Whether executing developer terminal commands, client agency proposals, content marketing audits, or corporate legal compliance, AIO.GEO provides deterministic workflows tailored to your specific technical and business requirements for modern production engineering teams.
Instant 1-command AST code fixes and CI score gates
$ npx @aio-geo/cli@0.3.4 audit example.com0-100 client health rating + $5k retainer pitch reports
Illustrative sample only. Live scores come from real audits.
Competitor LLM citation analyzer + answer-first H2 maps
DEMO-style illustration. Live mindshare requires probe fuel.
1-click AI copyright and opt-out disclosures
Full surface map: /paths
Protocol FAQ
Infrastructure, not prompt hacks. Receipts, not sampled dashboards. Free CLI, paid control room.
SEO and AEO score human search visibility on Google. AIO.GEO scores whether autonomous agents — Claude Code, Cursor, ChatGPT, and similar — can programmatically read your site, discover auth, and act. We measure structure, bot access, JSON-LD, RFC 9727 catalogs, MCP cards, and receipts. We do not sell LLM rankings. Nobody can measure those honestly.
Prompt hacks rot when Grok, Claude, or Perplexity change a ranking pass. Open protocols do not. We audit infrastructure agents already require to find you: RFC 9727 API catalogs, SEP-2127 MCP server cards, DNS _index._agents, llms.txt, auth.md, and clean Markdown. When a model updates, those endpoints stay live and discoverable. That is algorithm immunity: protocol layer, not prompt theater.
It is a proxy model, not measured revenue. Dual-axis scores two facts: Cognitive (can AI read you) and Kinetic (can AI act). The optional leakage number is score deficit × sector CPC prior × 10, with a 1.5× kinetic penalty when actionability is under 50. We hide it behind “Show proxy model” because it is agency math, not Search Console and not ChatGPT traffic. The product is the 0–100 score, the kinetic gap line, and the sealed receipt.
GSC AI reports are sampled, delayed, and noisy. AIO.GEO does not wait on them. Rescore is instant: line-by-line diffs for llms.txt, api-catalog, and server-card.json; kinetic pillar flips (for example 25 → 100 when MCP discovery lands); and a verified drop of the agent refusal line — before anything touches production. Dry-run first. HMAC receipt after. That is a deterministic receipt, not a sampled dashboard.
Open-core. The CLI (npx @aio-geo/cli@0.3.4), MCP server (@aio-geo/mcp-server), and single-domain audits are free. Pro is $49/mo for deep scans and fix packs. Agency is $199/mo for the multi-client control room — built so agencies resell AI Search Readiness retainers to high-CPC verticals (SaaS, legal, medical). You do not need enterprise to run an audit.
Method ·
AI Search Readiness is a deterministic 0 to 100 score that measures whether generative answer engines can crawl, parse, extract, and trust your public pages. It combines five weighted pillars: technical access for AI bots (25%), answer-first content structure (25%), schema markup completeness (20%), authority anchors and citations (20%), and presentation quality including alt text and semantic HTML (10%). It is not an LLM ranking product and never invents ChatGPT share of voice. Agencies use the score to prioritize repairs, ship dry-run fix packs, rescore after deploy, and attach sealed before-and-after receipts clients can verify.
The closed loop is measure, fix, apply, rescore. First, a free structural audit crawls your domain with SSRF-safe fetch, scores geo-heuristic-v3.5 pillars, and ranks lethal gaps such as blocked GPTBot rules or missing Organization schema. Second, a dry-run fix pack proposes robots.txt, JSON-LD graphs, and answer-first headers without writing your production site. Third, operators apply via CLI, MCP, or GitHub PR with tokens you control. Fourth, an identical methodology rescore produces a delta; optional HMAC seals prove the window when causal data exists. Every write path defaults to dry-run so CMOs review diffs before any live mutation ships.
Classic SEO optimizes for web ranking algorithms that weigh backlinks, keywords, and click models. Answer engines instead sample extractable HTML text, structured entities, bot access policies, and co-citation graphs across third-party sources. A site can rank well in Google while GPTBot is disallowed, JSON-LD is empty, or pages hide meaning behind thin marketing copy. AIO.GEO measures the structural surface those crawlers actually see, then produces machine-readable patches. For live multi-engine citation share you still need budgeted probes: readiness alone is necessary, not sufficient, for visibility outcomes.
CMOs receive plain-language readiness scores, lethal gaps, and competitor structure teardowns. Agencies receive multi-client tables, white-label digests, CLI and MCP apply paths, and optional sealed proof URLs. Product surfaces include free homepage audits, dashboard fix packs, operator CLI at /cli, and docs that refuse fake ranking claims.
Can models crawl, extract, and trust your pages? Five weighted pillars with plain-English fixes across 100 points of structural evidence.
See which third-party sites models learn from so you know where to earn placement without inventing answer-engine rankings.
Optional paid path: probe multiple answer engines under a hard spend cap. Always labeled, budget-gated, never silent cost overruns.
After fixes, a signed receipt shows whether readiness improved versus controls. Integrity of payload, not invented rankings.
Client table, seats, digests, and export built for five to fifteen thousand dollar monthly GEO retainers with white-label reports.
PDF, HTML, and digests your account team can send without explaining Greek variables or stage cosplay from internal engineering docs.
Know what AI sees
Start with a free structural audit. Graduate to dry-run fix packs, operator CLI, agency digests, and sealed before-and-after receipts when you are ready. Pick a path if you already know your role, or run the homepage scan for a baseline score.
Trust strip
We do not sell LLM rankings. We sell structure, accessibility, and receipts.