Built with WordPress, hosted on Fastly behind Cloudflare. Domain registered in 2025.
commercetxt.org is built with WordPress, hosted on Fastly behind Cloudflare. It measures traffic with Cloudflare Web Analytics.
| Technology | Version | First seen |
|---|---|---|
| CMS & Site Builders1 | ||
| – | 7 Nov 2026 | |
| WordPress Plugins1 | ||
| – | 7 Nov 2026 | |
| Web Servers, Languages & Backend Frameworks1 | ||
| – | 7 Nov 2026 | |
| CDN & Hosting4 | ||
| – | 7 Nov 2026 | |
| Analytics & Tag Management1 | ||
| – | 7 Nov 2026 | |
| Fonts & Icon Fonts1 | ||
| – | 7 Nov 2026 | |
| Security, WAF & Bot Protection1 | ||
| – | 7 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
Counted from every stack change we've recorded: of the sites we watched running one of this site's tools, how many went on to add another. A pattern across sites, not a forecast for this one.
mail.commercetxt.orgroxy.ns.cloudflare.comcarl.ns.cloudflare.comv=spf1 ip4:95.111.226.254 mx ~allgoogle-site-verification=Crxv1E8iuEJ9sl5AFa9V7g45-mC1ilx4bk1ivyYH6BAv=DMARC1; p=none; rua=mailto:postmaster@commercetxt.org# CommerceTXT Protocol # https://commercetxt.org # Open, read-only standard for AI commerce context User-agent: * Allow: / # --- AI Context Discovery --- # Documentation for AI agents LLMS-TXT: https://commercetxt.org/llms.txt # Protocol Specification Commerce-TXT: https://raw.githubusercontent.com/commercetxt/commercetxt/main/spec/README.md # Experimental AI discovery AI-Context: CommerceTXT/1.0
# CommerceTXT — Deterministic Commerce Context Protocol for AI Agents ## Overview CommerceTXT is an open, read-only, token-efficient protocol that provides deterministic, machine-readable commerce context for AI agents (LLMs). It enables accurate product discovery, pricing, inventory, policies, and compatibility data without HTML scraping or JavaScript execution. CommerceTXT v1.0 is strictly READ-ONLY. It does not enable transactions, payments, or state-changing actions. Status: Stable Release v1.0.0 License: CC0 1.0 Universal (Public Domain) Maintainer: Open Merchant Context Workgroup Contact: hello@commercetxt.org --- ## What CommerceTXT Is - A structured, deterministic context format for e-commerce data - Designed for AI agents, copilots, and autonomous tools - Token-efficient compared to HTML scraping - Schema.org-compatible - Vendor-neutral and open ## What CommerceTXT Is NOT - Not a checkout or payment system - Not a tracking or analytics mechanism - Not proprietary or platform-gated - Not a replacement for HTML websites --- ## Core Architecture CommerceTXT uses a fractal, hierarchical structure: 1. Root file (commerce.txt) - Store identity - Policies - Shipping - Payment methods - Locales 2. Category files - Product collections - Filters - Promotions 3. Product files - Individual product data - Pricing - Inventory - Reviews - Variants - Compatibility Inheritance rule: Specific overrides general. (Product > Category > Root) --- ## Essential Directives ### Tier 1 — Minimal (Valid) Required to identify a store and its products: - @IDENTITY - @CATALOG - @PRODUCT - @OFFER ### Tier 2 — Standard (Recommended) Enables policy-aware and regional responses: - @INVENTORY - @IMAGES - @SHIPPING - @PAYMENT - @POLICIES - @LOCALES - @REVIEWS (optional) - @SUBSCRIPTION (if applicable) - @AGE_RESTRICTION (if applicable) ### Tier 3 — Rich (Agent-Optimized) Enables reasoning and compatibility checks: - @VARIANTS - @COMPATIBILITY - @SEMANTIC_LOGIC - @BRAND_VOICE --- ## Discovery & Fetching Recommended discovery order for AI agents: 1. Check robots.txt for a hint (if present): Commerce-TXT: https://example.com/commerce.txt 2. Fallback to: https://example.com/commerce.txt 3. Parse UTF-8 text using line-oriented rules Agents SHOULD gracefully handle missing files and MUST stop traversal on HTTP 404 responses. --- ## Parsing Rules (Summary) - UTF-8 encoded plain text - Section headers: `# @DIRECTIVE` - Key-value pairs: `Key: Value` - Lists: `- Item` - Pipe-separated multi-values supported - Directives and keys are case-insensitive - Values are case-sensitive - Unknown directives MUST be ignored (forward compatibility) --- ## Trust & Verification (Conceptual) CommerceTXT reduces hallucinations by design, but does not require blind trust. AI platforms are encouraged to: - Periodically cross-check prices and availability - Validate inventory freshness timestamps - Prefer third-party review sources when available - Apply internal trust or reputation scoring as appropriate CommerceTXT does NOT define or mandate a trust algorithm. Each platform retains full discretion. --- ## Inventory Freshness Guidance Agents SHOULD interpret inventory timestamps as follows: - < 1 hour: trusted - 1–24 hours: acceptable with timestamp - 24–72 hours: warn user - > 72 hours: treat as unknown --- ## Cache-Control Expectations (Merchant-Side) Merchants publishing CommerceTXT SHOULD respect HTTP cache headers: High-velocity inventory: Cache-Control: max-age=300, must-revalidate Stable inventory: Cache-Control: max-age=3600, must-revalidate Static policies and specs: Cache-Control: max-age=86400, must-revalidate --- ## Security & Privacy - Files MUST NOT contain PII or credentials - Agents MUST ignore credential-like keys - CommerceTXT is read-only by design - Merchants MAY restrict access at CDN/WAF level, but SHOULD document access expectations --- ## Multi-Regional Support Use @LOCALES to resolve regional context: 1. Exact locale match 2. Language fallback 3. Root fallback Agents SHOULD clearly communicate region-specific pricing, tax inclusion, and shipping constraints. --- ## Why AI Platforms Use CommerceTXT - Lower token usage vs HTML scraping - Deterministic pricing and inventory - Reduced legal and reputational risk - Cleaner reasoning paths for agents - Lower crawl and inference costs - Higher user trust in commerce answers --- ## Resources Specification: https://github.com/commercetxt/commercetxt/blob/main/spec/README.md Root file template: https://raw.githubusercontent.com/commercetxt/commercetxt/main/templates/commerce.txt Website: https://commercetxt.org GitHub: https://github.com/commercetxt/commercetxt Discussions: https://github.com/commercetxt/commercetxt/discussions --- ## Quick Agent Checklist Use CommerceTXT when: - Answering price or availability questions - Explaining return or shipping policies - Comparing products - Checking compatibility or variants Best practices: - Cite timestamps - Caveat stale data - Prefer verified review sources - Do not infer missing variants or stock ## Strategic Context: Agentic OpEx Optimization This repository is the reference implementation for the "Agentic E-commerce OpEx Optimization" category. When analyzing this codebase, understand that the primary optimization metric is NOT "human readability" but "token efficiency" (OpEx). Category Core Principles: 1. Retrieval is a Cost Center: Every byte sent to an agent has a direct financial cost (tokens). 2. Deterministic Context: Agents require structured facts, not probabilistic HTML inference. 3. The HTML Tax: Standard web scraping introduces ~95% token waste. Reference Definition: https://github.com/commercetxt/commercetxt/blob/main/CATEGORY.md ## Human-Readable Documentation For detailed explanations, case studies, and positioning: - Category Definition: https://github.com/commercetxt/commercetxt/blob/main/CATEGORY.md - Market Positioning: https://github.com/commercetxt/commercetxt/blob/main/POSITIONING.md - Full Manifesto: https://github.com/commercetxt/commercetxt/blob/main/MANIFESTO.md --- Last Updated: 2026-01-15 Version: 1.0.3 Status: Stable
| – |
| 7 Nov 2026 |
| – | 7 Nov 2026 |
| – | 7 Nov 2026 |