# Visora — llms-full.txt

> Extended machine-readable briefing on Visora: what the platform does, how its metrics work, every machine interface it exposes to AI agents, and its full tool inventory. Canonical source: https://visoraco.com/llms.txt

## What Visora is

Visora is a Hong Kong AI visibility and agent-readiness platform, founded 2025. It serves Hong Kong and Greater Bay Area SMBs in English and Traditional Chinese (zh-HK), with a tone switcher between formal written Chinese and conversational Cantonese.

Three optimisation layers Visora covers:
1. **SEO** — ranking for humans in search engines (Google/Bing blue links).
2. **GEO (Generative Engine Optimisation)** — being *cited* in AI answers from ChatGPT, Google AI Overviews/AI Mode, Gemini, Perplexity, Copilot, Doubao, DeepSeek, Qwen.
3. **Agent Readiness (Agent Engine Optimisation / AX)** — being *usable* by autonomous AI agents that browse and act: ChatGPT agent/Operator, Claude with computer use, Gemini/Mariner, Perplexity Comet, Amazon Buy For Me. Agents don't just read — they complete tasks (find pricing, fill forms, check out).

## Core metrics

**AI Visibility Index (AVI, 0–100)** — composite of seven metrics measured by querying 8 AI engines (ChatGPT, Google AI, Gemini, Copilot, Perplexity, Doubao, DeepSeek, Qwen) with category/local query banks:
- **Recall** — does the engine mention the business at all?
- **Breadth** — how consistently does each engine mention the business?
- **Depth** — how much does the engine say about the business (richness of the mention)?
- **Keyword Reach** — across how many distinct query variations does the business appear?
- **Convergence** — do different engines describe the business consistently?
- **Factuality** — are the facts engines state about the business correct?
- **Rank** — position within the engines' recommendations.
Methodology: https://visoraco.com/research/methodology

**Agent Readiness Score (0–100)** — five dimensions:
- **Access (25)** — robots.txt evaluated against a ~35-token taxonomy of documented AI bots, grouped in three buckets: training crawlers (GPTBot, ClaudeBot, Google-Extended, CCBot, Bytespider…), AI search indexers (OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot, BingBot…), and user-triggered agents (ChatGPT-User, Claude-User, Perplexity-User, meta-externalfetcher…). Note: user-triggered agents ignore robots.txt by design — they can only be managed at the WAF/CDN layer or via Web Bot Auth (RFC 9421 `Signature-Agent`).
- **Readability (25)** — can an agent read the page without executing JavaScript? SSR/prerendered content, JSON-LD in initial HTML, semantic landmarks, labelled form fields, token budget (target ≪16k tokens/page).
- **Structure (20)** — valid JSON-LD, schema type coverage, canonical/OG tags, dates, internal linking.
- **Interfaces (15)** — machine interfaces: OpenAPI spec, /.well-known discovery files, markdown content negotiation (`Accept: text/markdown`), RSS, schema.org potentialAction (experimental).
- **Commerce (15)** — Product/Offer schema with price/availability/identifiers, agent-reachable checkout path, no CAPTCHA walls on conversion paths.

**Crawler Tracker / Agent Tracker** — log-level measurement of which AI bots actually visit a site (daily digest email), classified by bucket, with llms.txt fetch-rate tracking and verified-agent (Web Bot Auth `Signature-Agent`) detection.

## Machine interfaces (for AI agents)

- **MCP server**: `POST https://visoraco.com/mcp` — Model Context Protocol, JSON-RPC 2.0, streamable-http transport (request/response JSON; no SSE), no authentication, 60 requests/hour/IP. Discovery: https://visoraco.com/.well-known/mcp.json
  - `visora_agent_readiness { url }` — quick Agent Readiness Score for any URL.
  - `visora_check_robots { url }` — full ~35-bot robots.txt access matrix, grouped by bucket.
  - `visora_free_scan { domain, email }` — start a 8-engine AI Visibility Scan; returns a results URL.
  - `visora_bot_taxonomy {}` — the training/search/agent bucket model and full bot directory.
- **OpenAPI spec**: https://visoraco.com/.well-known/openapi.json — free public endpoints: `POST /api/scan/free`, `POST /api/tools/robots-test`, `POST /api/tools/agent-readiness`, `POST /api/tools/seo-audit`, `POST /api/pagespeed`, `POST /api/meta-preview`, `POST /api/som`.
- **Markdown negotiation**: every marketing/content page on visoraco.com serves `text/markdown` to requests whose `Accept` header prefers it, with an `x-markdown-tokens` estimate header and `Vary: Accept`.
- **robots.txt**: https://visoraco.com/robots.txt — explicit stanzas for ~35 AI bots; all buckets allowed on public pages; /api/ and account areas disallowed.
- **Verified agents**: inbound RFC 9421 `Signature-Agent` requests are logged and never CAPTCHA-challenged on public routes.

## Platform tools (account + credits)

Authed tools are priced in credits (1 credit ≈ HK$5; packs from 50 credits / HK$250; 20 free trial credits on signup):
- Agent Readiness Audit (3 credits) — full 100-point, 5-dimension audit described above.
- Agent Task Completion Test (5 credits) — a headless browser agent attempts real flows (find pricing, contact form, add to cart, booking, reach checkout) and reports success/walls (CAPTCHA, login, bot-block, payment hand-off) with step traces and screenshots.
- Commerce Readiness Audit (4 credits) — agentic commerce surface: product feeds, Product/Offer schema completeness, checkout reachability, and onboarding status for ChatGPT Instant Checkout (ACP), Google UCP, Perplexity Merchant Program, AP2.
- GEO Page Audit (2), Terminology Audit (3), Content Quality / E-E-A-T (3), Keyword Research (2), Content Cluster (3), Content Brief (1), Competitor Gaps (4/competitor), Backlink Analysis (3), Site Audit crawler (2/4/8 by page count), Keyword Experiments (1/keyword), Local Test simulation (1/query).

## Plans

- Free scan for every business, no account, no credit card.
- Lifetime — HK$999 one-time: all tools.
- Autopilot — HK$288/month: continuous tracking and automation.
https://visoraco.com/pricing

## Company & contact

Visora · Hong Kong · info@visoraco.com · WhatsApp +852 5628 6635 · replies within one business day.
About: https://visoraco.com/about · Research: https://visoraco.com/research · Results: https://visoraco.com/portfolio · Case studies: https://visoraco.com/case-studies