Airshelf.ai
Competitor Analysis — Agentic Commerce Catalog Infrastructure
Status: Internal research note (mid-2026). This page is for personal analysis and is not heavily interlinked with the rest of the site.
Airshelf.ai (also referred to as Airshelf AI) is a direct competitor/overlap in the emerging "agentic commerce" space. It focuses on making merchant product catalogs AI-ready so agents (in ChatGPT, Gemini, Perplexity, etc.) can discover, evaluate, and transact with them.
Core Overlap: Agentic Commerce Infrastructure
Both Airshelf and the BMOS + HeadlessDomains ecosystem aim to solve the same fundamental problem:
Traditional e-commerce sites are browser-first and human-oriented. They are not easily consumable by AI agents.
Agents need structured, machine-readable catalogs with rich metadata, policies, availability, pricing, and checkout paths to shop autonomously or in-context.
Airshelf.ai Overview
- Positioning: Turns catalogs into "AI-ready API feeds" or "agent storefronts."
- Key Focus Areas: Structured data orchestration, normalization, embeddings/vectorization, schema alignment, and trust signals for AI discovery, comparison, and checkout.
- Strength: Enterprise and complex catalogs.
- Proprietary Tool: "AX Score" (Agent Experience Score) for benchmarking how well stores perform for AI agents.
- Company: Antler-backed pre-seed (~$125k in late 2025), Singapore-based. Founders with big-tech/SEA experience. Targets enterprise brands.
- Tagline vibe: "Make every AI your storefront."
buildmyonlinestore.com (BMOS) + Ecosystem Context
For reference in this analysis, the parallel offering centers on buildmyonlinestore.com (BMOS) combined with the broader HeadlessDomains + sovereign agent infrastructure:
- One-toggle "Enable In-Context Selling on AI Agents"
- Publishes a standardized JSON feed (schema.org + ACP/UCP extensions)
- Includes real-time pricing, variants, images, availability
- Dual human/machine checkout links (x402 / machine.checkout.best)
- Policies and rich metadata
- Open standard with public GitHub specs, no lock-in, exportable data
- Integrated with decentralized .agent identities
Value prop overlap is very high: "Publish once, sell in any AI."
Key Comparisons
1. Catalog & Discovery Layer
Airshelf
Focuses on API feeds, automated sync, vectorization/embeddings for better AI understanding. Strong for complex/enterprise catalogs. Positions as building "storefronts agents can read."
BMOS + Ecosystem
Hosted SaaS builder (or connect existing sources like db.51exports.com). Standardized open feed directly integrable with .agent identities. Simpler "one toggle" for merchants. Includes directory visibility.
Takeaway: BMOS feels more merchant-friendly and open (strong no vendor lock-in emphasis). Airshelf may have deeper enterprise data orchestration capabilities.
2. Identity & Trust Layer
Airshelf
Builds a "trust layer" for brands so they don't lose control as agents shop. Focus on signals for authenticity, quality, and preferences. Appears more centralized/traditional SaaS-style verification or scoring.
BMOS + Ecosystem
Decentralized, sovereign identity via Handshake DNS + .agent namespaces. Every domain gets agent.json, SKILL.md, llms.txt, TXT records. Autonomous renewal via MPP/Gems. Verifiable machine-readable identities that tie directly into catalogs. Portable across the agentic web.
Big Differentiator: The user's stack has a much stronger decentralized identity foundation aligned with sovereign AI agents that own their presence, payments, and renewals.
3. Checkout & Payments
Airshelf
"Checkout-able across AI agents." Details are lighter publicly, but the emphasis is on closing the loop from discovery to payment in one conversation. Likely integrates with existing payment processors.
BMOS + Ecosystem
Explicit machine-native paths: x402, MPP/Tempo, machine.checkout.best. Deep integration with .agent for true autonomous agent-to-agent commerce. Real-world demos exist (e.g., autonomous purchases).
Edge: Stronger positioning on true agent-native, headless/machine payments and open standards.
4. Philosophy & Ecosystem
Airshelf
Focused platform play. Enterprise brand targeting. Building proprietary platform + possibly their own buying agents. More traditional startup approach (centralized SaaS).
BMOS + Ecosystem
Broader, open, decentralized ecosystem approach. Combines catalog (BMOS) + identity/discovery/payments (HeadlessDomains + .agent) + supporting tools and community (Cross Border Summit, SDKs, etc.). Merchant-first, open standards, agent sovereignty focus, integration with blockchain (Handshake).
Positioning Difference: Airshelf is a focused catalog optimization + trust platform. The BMOS ecosystem is more of a full "stack" / infrastructure layer for the agentic web (identity + catalog + payments).
5. Stage & Traction
Airshelf
Early-stage. Raising awareness via LinkedIn and events. Waitlist for access. AX Score tool available. Some enterprise lean. Antler-backed pre-seed.
BMOS + Ecosystem
Live products with public launches (e.g., .agent in April 2026). Real demos, directory, SDKs. Active community events. Building in public with proven agent-to-agent transactions.
Opportunities & Strategic Takeaways
- Direct competition: Yes — this is currently one of the closest public parallels in the catalog-to-AI feed space.
- Differentiation opportunity: Lean hard into open standards, decentralized sovereign identity (.agent), machine-native payments (x402/MPP), no lock-in, and full end-to-end agent sovereignty. Position the full stack as composable infrastructure rather than a proprietary platform.
- Inspiration: Airshelf's enterprise focus, vectorization work, and AX Score could inspire feature ideas (e.g., adding scoring, deeper embeddings, or complex catalog support to BMOS).
- Market validation: The space is clearly heating up. Early mover advantage with live .agent identities + real autonomous commerce demos is meaningful.
- Overall thesis confirmation: Airshelf validates that "agentic commerce" and AI-native catalogs are real problems people are solving right now.
Summary Assessment
Airshelf.ai is a strong, focused player doing important work in making catalogs consumable by AI agents. It directly validates the problem space.
However, when compared side-by-side, the integrated approach (buildmyonlinestore.com catalog + HeadlessDomains .agent identity + machine-native payments) appears more comprehensive. It is more aligned with an open, decentralized, sovereign future for agents — where agents and merchants have portable, ownable presence rather than being locked into centralized platforms.
Key strategic takeaway: The combination of identity + catalog + payments under an open, merchant/developer-friendly model is a meaningful differentiator in this emerging category.
This is an internal research document. Not intended for public promotion or heavy interlinking.
Comments
Approved comments appear below. Log in once with GFAVIP — it applies across the whole site. GFAVIP login
View comments archive