# Zeodyn > Agent commerce assessment platform. Can AI agents do business with you? Zeodyn™ is an AI commerce readiness platform built and operated by Virtual Factory Solutions Ltd. (trading as Zeodyn). The platform assesses whether AI agents — shopping assistants, autonomous buyers, comparison engines — can discover, evaluate, and transact with e-commerce websites. The primary assessment framework is the Agent Commerce Stack™, which evaluates websites across six weighted dimensions and produces a composite Zeodyn Score™ from 1 to 100. Zeodyn serves both as a marketing site (zeodyn.com) and as the live product delivery platform. The flagship product, the Zeodyn Scanner™, is embedded directly on the site and available to use for free. --- ## Products ### Zeodyn Scanner™ Free tool available at https://zeodyn.com/scan. Submit any URL and receive an instant Zeodyn Score™ — a composite metric from 1 to 100 — that measures how ready a website is for AI agent commerce. The scanner runs 54 automated sub-checks across six dimensions of the Agent Commerce Stack™ framework. Results include a full dimension breakdown, sub-check details, a radar chart visualisation, and actionable recommendations for improvement. Free tier scans the homepage only. Paid tiers unlock multi-page scanning, which discovers product pages via sitemap and structured data for a more representative assessment. Multi-page scanning measures product pages where commerce signals naturally live, giving a more representative assessment of agent readiness. ### Zeodyn Score™ The Zeodyn Score™ is the composite metric produced by the scanner. It ranges from 1 to 100 and is calculated using a weighted geometric mean of the six dimension scores, with fail gates for critical capabilities. The score places each site into one of five bands: - **Not Ready (1–24):** AI agents cannot discover or transact with the site. - **Limited (25–49):** Major gaps prevent agents from completing commerce interactions. - **Developing (50–69):** AI agents can discover the site but struggle with evaluation and transactions. - **Strong (70–89):** The site supports AI agent interactions well, with clear opportunities for improvement. - **Agent-Ready (90–100):** The site is fully accessible to AI agents across all dimensions. ### Zeodyn Monitor Subscription product for ongoing monitoring and alerts. Watched sites are re-scanned weekly, with score trend charts, email digests, custom threshold alerts, and webhook notifications when scores change. Available on Pro, Growth, and Enterprise tiers. --- ## Agent Commerce Stack™ — Six Dimensions The Agent Commerce Stack™ framework assesses websites across six weighted dimensions. Each dimension answers a specific question about AI agent readiness. ### 1. Discovery & Access **Question:** Can AI agents find and access your commerce capabilities? **Weight:** Very High Measures whether AI agents can locate the site through standard discovery mechanisms — robots.txt directives, meta tags, sitemaps, protocol discovery files (llms.txt, agent.json, MCP manifest, UCP profile), and HTTP accessibility. Sites that block or fail to expose themselves to agents score poorly here. ### 2. Structured Data **Question:** Can AI agents understand your products in machine-readable form? **Weight:** Very High Assesses the presence, correctness, and completeness of structured data markup — Schema.org Product, Offer, Organization, BreadcrumbList, and other types that allow agents to parse product information without relying on visual page rendering. ### 3. Commerce Data **Question:** Can AI agents trust your operational data for transactional decisions? **Weight:** High Evaluates whether key commerce signals are machine-readable: pricing, currency, availability, shipping information, return policies, payment methods, and offers. Agents need reliable, consistent commerce data to make purchasing decisions on behalf of users. ### 4. Protocol Support **Question:** Can AI agents programmatically transact with your commerce infrastructure? **Weight:** High Checks for support of agent commerce protocols — the Universal Commerce Protocol (UCP), Agentic Commerce Protocol (ACP), Model Context Protocol (MCP), A2A, OpenAPI specifications, and other machine-to-machine interfaces that enable programmatic interaction. ### 5. Security & Trust **Question:** Can AI agents verify your legitimacy and operate safely? **Weight:** Moderate Verifies HTTPS configuration, certificate validity, security headers (HSTS, CSP, X-Frame-Options), content security policies, and trust signals that allow agents to assess whether a site is safe to transact with. ### 6. Technical Performance **Question:** Can AI agents parse your pages quickly and efficiently? **Weight:** Moderate Measures page load speed, response times, content size, compression, and rendering characteristics. AI agents process pages programmatically and are sensitive to slow or bloated responses that increase latency and cost. --- ## API Overview Zeodyn provides a REST API with 15 primary endpoint groups. Authentication uses API keys passed via the X-API-Key header (prefix `zd_live_`). All responses are JSON. **Endpoint groups:** - **Account & Auth:** Registration (magic link), account info, subscription management, billing portal, API key CRUD. - **Scanning:** Submit URLs for scanning, retrieve results by ID, list scan history. - **Batch Scanning:** Submit up to 50 URLs per request (Growth tier), check batch status. - **Watched Sites:** Add, list, update, and remove monitored sites with score history. - **Webhooks:** Register HTTPS webhook endpoints for score change, scan complete, threshold alert, and batch complete events (Pro+). - **Thresholds:** Custom score alert thresholds per watched site (Growth+). Full OpenAPI 3.1 specification: https://zeodyn.com/openapi.json API documentation: https://zeodyn.com/docs/api --- ## Pricing (March 2026) All prices exclude VAT. - **Anonymous Free:** £0 — 3 scans per day per IP, homepage-only scan, full score breakdown and recommendations, shareable report link. - **Signed-Up Free:** £0 — 10 scans per day, 3 watched sites, 90-day scan history, 100 API calls per month, homepage-only scan. - **Scanner Pro:** £19/month — Unlimited scans, 2-page scanning (homepage + 1 product page), 10 watched sites with weekly re-scans, score trend charts, PDF certificate export, AI-powered recommendations, 1,000 API calls per month, 1 webhook. - **Scanner Growth:** £49/month — Everything in Pro, multi-page scanning (up to 5 pages: homepage + up to 4 product pages), unlimited API calls, batch scanning (up to 500 URLs), team workspace (up to 5 members), competitive benchmarking, multiple webhooks, custom score threshold alerts. - **Scanner Enterprise:** Custom pricing — Custom page depth, custom dimension weights, white-label portal, SSO (SAML/OIDC), dedicated scanning infrastructure, SLA, bulk data export. Machine-readable pricing: GET https://zeodyn.com/api/pricing --- ## Protocol & Discovery Files - **/llms.txt** — LLM-readable summary of the site, products, APIs, and documentation links. - **/llms-full.txt** — Extended LLM context with full descriptive content inline (this file). - **/agent.json** — Agent discovery metadata with full API catalogue, documentation links, and protocol references. - **/openapi.json** — OpenAPI 3.1 specification with 40+ endpoints, request/response schemas, and authentication details. - **/.well-known/ai-plugin.json** — ChatGPT/OpenAI plugin discovery manifest. - **/.well-known/security.txt** — Security contact and vulnerability disclosure policy. - **/.well-known/ucp** — Universal Commerce Protocol capability declaration with transport bindings. - **/.well-known/agent-card.json** — Google Agent-to-Agent protocol agent card with skills and runtime declaration. - **/.well-known/mcp.json** — Model Context Protocol discovery manifest with tool schemas and authentication info. - **/.well-known/acp/config.json** — Agentic Commerce Protocol discovery manifest with product catalogue, payment provider, and checkout endpoints. - **/robots.txt** — Crawler directives. All major AI agents explicitly allowed. - **/sitemap.xml** — Dynamic XML sitemap covering marketing pages, documentation, blog posts, and protocol files. --- ## Methodology The Zeodyn Score™ is produced by a fully automated scanning pipeline. The scanner sends HTTP requests from UK cloud infrastructure (Azure, UK South) and analyses the responses across 54 sub-checks grouped into six weighted dimensions. Each sub-check produces a normalised score, and these are aggregated within their dimension. The six dimension scores are then combined using a weighted geometric mean to produce the final composite score. Fail gates ensure that critical capabilities cannot be ignored. If a site scores below a minimum threshold in any dimension, the fail gate mechanism caps the overall score regardless of strength in other areas. This prevents a site from achieving a high composite score while completely lacking, for example, structured data or commerce data. The scoring methodology has been empirically validated and calibrated against more than 3,500 sites across 17 markets and 3 verticals (fashion, grocery, electronics). Principal component analysis confirms the six-dimension structure, and weight sensitivity analysis shows strong rank-order stability (Spearman rho > 0.96 across all perturbations). The full methodology is published at https://zeodyn.com/methodology. Version: v1.0.1 (March 2026). --- ## Documentation (24 pages) - [Documentation Home](https://zeodyn.com/docs): Overview and navigation for all documentation. - [Getting Started](https://zeodyn.com/docs/getting-started): First scan guide, account setup, tier comparison. - [Scanner Features](https://zeodyn.com/docs/scanner): Overview of scanner capabilities and dimension breakdown. - [Watched Sites](https://zeodyn.com/docs/scanner/watched-sites): Monitor sites for score changes with weekly re-scans. - [Trend Charts](https://zeodyn.com/docs/scanner/trend-charts): Score history visualisation over time per dimension. - [Benchmarking](https://zeodyn.com/docs/scanner/benchmarking): Compare scores against competitors in your vertical. - [Batch Scanning](https://zeodyn.com/docs/scanner/batch-scanning): Scan up to 50 URLs per request via API. - [PDF Export](https://zeodyn.com/docs/scanner/pdf-export): Generate downloadable score certificates. - [AI Recommendations](https://zeodyn.com/docs/scanner/ai-recommendations): AI-enhanced improvement guidance tailored to your results. - [Thresholds](https://zeodyn.com/docs/scanner/thresholds): Custom score alert thresholds per watched site. - [API Reference](https://zeodyn.com/docs/api): Authentication, rate limits, error codes, and all endpoint schemas. - [Scan API](https://zeodyn.com/docs/api/scan): Submit URLs for scanning and retrieve results. - [Results API](https://zeodyn.com/docs/api/results): Retrieve scan results by ID with full dimension breakdown. - [Batch API](https://zeodyn.com/docs/api/batch): Batch scanning endpoints for high-volume assessments. - [Account API](https://zeodyn.com/docs/api/account): Account management, subscription, billing, and API keys. - [Watched Sites API](https://zeodyn.com/docs/api/watched-sites): CRUD for watched sites with SSRF validation. - [Webhooks API](https://zeodyn.com/docs/api/webhooks-api): Webhook CRUD, test delivery, and HMAC-SHA256 verification. - [Thresholds API](https://zeodyn.com/docs/api/thresholds): Score threshold management for Growth tier and above. - [Webhooks Guide](https://zeodyn.com/docs/webhooks): Setup guide, event types, payload schemas, and signature verification. - [Teams](https://zeodyn.com/docs/teams): Team workspaces, roles, and shared data across members. - [MCP Server](https://zeodyn.com/docs/mcp): Model Context Protocol integration — endpoint, authentication, 8 tools, JSON-RPC methods. - [A2A Endpoint](https://zeodyn.com/docs/a2a): Agent-to-Agent integration — natural language tasks, intent detection, task lifecycle. - [ACP Checkout](https://zeodyn.com/docs/acp): Agentic Commerce Protocol — checkout lifecycle, endpoints, SharedPaymentTokens, digital fulfilment. - [FAQ](https://zeodyn.com/docs/faq): Common questions and answers about the scanner, scoring, and platform. --- ## Blog - [What Is Agent Commerce?](https://zeodyn.com/blog/what-is-agent-commerce): A definitive guide to AI agents that discover, evaluate, and buy on behalf of consumers. - [Zeodyn Score Explained](https://zeodyn.com/blog/zeodyn-score-explained): How the Zeodyn Score™ works — dimensions, weights, geometric aggregation, fail gates, and what your score means. - [Shopify AI Agent Readiness](https://zeodyn.com/blog/shopify-ai-agent-readiness): Seven practical steps to prepare your Shopify store for AI agent commerce. - [UCP vs ACP Protocol Guide](https://zeodyn.com/blog/ucp-vs-acp-protocol-guide): Google told retailers to get ready for AI shoppers — a guide to the UCP and ACP protocols. RSS feed: https://zeodyn.com/feed.xml --- ## MCP Integration Zeodyn exposes a Model Context Protocol (MCP) server for native AI agent integration. The server implements JSON-RPC 2.0 over Streamable HTTP transport and provides 8 tools for scanning, monitoring, and account management. ### Endpoints - **Generic:** POST https://zeodyn.com/api/mcp/scanner - **Claude-optimised:** POST https://zeodyn.com/api/mcp/claude/scanner - **Health check:** GET https://zeodyn.com/api/mcp/health ### Authentication API key via `X-API-Key` header (prefix `zd_live_`). Optional — unauthenticated access is available with rate limits. Register via POST https://zeodyn.com/api/v1/auth/register with `{ "email": "you@example.com" }`. ### Rate Limits - Anonymous: 200 unique domains/day - Authenticated: Per-tier limits (see https://zeodyn.com/pricing) ### Tools (8) **1. scan_website** — Scan a URL for AI agent commerce readiness. Returns a Zeodyn Score™ (1–100) with dimension breakdowns. Free tier scans homepage only. Pro scans 2 pages, Growth scans up to 5 pages with automatic product page discovery. - Parameters: `url` (string, required) — The URL to scan. **2. get_scan_results** — Retrieve a previous scan result by its UUID. - Parameters: `scan_id` (string, required) — The scan result UUID. **3. list_scans** — List scan history with pagination. Requires authentication. - Parameters: `page` (number, optional, default 1) — Page number. `limit` (number, optional, default 20, max 100) — Results per page. **4. get_plans** — List available Zeodyn subscription plans with pricing, features, and limits. - Parameters: None. **5. get_account** — Get account details, tier, and usage stats. Requires authentication. - Parameters: None. **6. list_watched_sites** — List watched sites with score history. Requires authentication. - Parameters: None. **7. add_watched_site** — Add a site to your watchlist for ongoing monitoring. Requires authentication. - Parameters: `url` (string, required) — The URL to watch. `label` (string, optional) — Label for the site. **8. compare_sites** — Compare Zeodyn Scores across multiple URLs for competitive benchmarking. - Parameters: `urls` (array of strings, required) — URLs to compare (2–5). --- ## Integrations Zeodyn Scanner is available on: - ChatGPT GPT Store: https://chatgpt.com/g/g-69b5a0808aa88191b8d081c385a461b4-zeodyn-scanner-ai-commerce-readiness - Claude Code via MCP: claude mcp add --transport http zeodyn https://zeodyn.com/api/mcp/scanner - Chrome Extension: https://chromewebstore.google.com/detail/zeodyn-scanner-ai-commerce-readiness/ameigmgemknkimgdfiafijedcdodbigp - Web: https://zeodyn.com/scan - API: https://zeodyn.com/docs/api --- ## Contact - General enquiries: hello@zeodyn.com - Security issues: security@zeodyn.com - Contact form: https://zeodyn.com/contact --- © 2026 Virtual Factory Solutions Ltd. All rights reserved worldwide. Zeodyn™, Zeodyn Scanner™, Zeodyn Score™, and Agent Commerce Stack™ are trademarks of Virtual Factory Solutions Ltd.